AN ABSTRACT OF THE THESIS OF
Umur Onal for the degree of Master of Science in Fisheries Sicence presented on August
28, 1997. Title: Growth and Survival of Zebrafish, Brachydanio rerio, Larvae Fed on
Two Types of Microparticulate Diets.
Redacted for Privacy
Abstract approved:
Christopher J.
Effects of diet and container type on growth and survival of altricial zebrafish,
Brachydanio rerio, larvae were determined. Microfeast® L- 10 (MF) supported growth
and survival of zebrafish larvae during the first 10-12 days. Best results were obtained
with larvae fed on a combination of MF and Artemia nauplii in Imhoff, 1.2 1 cones.
The characteristics of two types of microparticulate diets were determined. MF
was encapsulated within cross-linked protein walled capsules (CLPWC) or gelatinalginate beads (GAB). Retention efficiencies were determined by encapsulating a nontoxic dye (Poly-R 478) within CLPWC or GAB. Retention efficiencies of both particle
types suggested that high molecular weight, water soluble nutrients could be delivered to
freshwater fish larvae without major losses. Acceptability experiments included
determination of gut fullness and feeding incidence of larvae using a computer-aided
image analysis system. Acceptability of GAB by first feeding zebrafish larvae was
significantly greater than that for CLPWC. Acceptability of CLPWC increased in larger
size larvae and was similar to that for GAB. It was determined that both particle types
were digested by first feeding zebrafish larvae.
Diet size selection of first feeding and 15 day-old zebrafish larvae was
determined with CLPWC using a model developed to account for both settlement of
capsules and ingestion by larvae. Results indicated that while first feeding larvae
preferred 2 1-45 tm capsules, 15 day-old larvae preferred capsules in the 46-75 tm size
range.
Growth experiments with CLPWC showed that up to 40% substitution of Arteinia
nauplii could be accomplished without reduced growth and survival of zebrafish larvae
after a feeding period of 8 days. Twenty percent substitution of Arternia nauplii could be
achieved with GAB without reduced growth and survival. Higher substitutions of
Anemia with either particle types resulted in inferior growth and survival of zebrafish
larvae.
Growth and Survival of Zebrafish,
Brachydanio rerio,
Fed on Two Types of Microparticulate Diets
by
Umur Onal
A THESIS
Submitted to
Oregon State University
In partial fulfillment of
the requirements for the
degree of
Master of Science
Completed August 28, 1997
Commencement June 1998
Larvae
Master of Science thesis of Umur Onal presented on August 28. 1997
APPROVED:
Redacted for Privacy
Major Professor, representing Fisheries
Redacted for Privacy
Head of Department of Fisheries and Wildlife
Redacted for Privacy
Dean of Graduate'Echool
I understand that my thesis will become part of the permenant Collection of Oregon State
University libraries. My signature below authorizes release of my thesis to any reader
upon request.
Redacted for Privacy
Umur Onal, Author
ACKNOWLEDGMENT
First and foremost, I would like to thank my wife for her patience and endurance
during the long and rainy nights of Newport. Without her help it would be much harder to
finish this study on time.
I would also like to thank Dr. Chris Langdon for his supervision of this study,
particularly his endless editing will of my drafts. I also would like to thank Dr. Virginia
Lesser for her help and suggestions during our monthly statistical meetings.
I would like to thank my friends, Mehmet Emin Alpay and Tolga Acar for their
help. Special thanks to David Jacobsen who showed and kindly helped me make gelatin-
alinate heads. I would also like to thank Dr. Mike Rust for his valuable comments
through e-mail.
TABLE OF CONTENTS
Page
INTRODUCTION
.1
Aspects of Larval Fish Development, Feeding
andDigestion Physiology ........................................................................
Development of Microparticulate Diets and
Their Use for Fish Larvae ........................................................................ 5
MATERIALS AND METHODS ............................................................................ 9
GeneralMethods ...................................................................................... 9
Culture system for broodstock ..................................................... 9
Breeding ....................................................................................... 9
Larval measurements ................................................................... 10
Waterquality ................................................................................ 11
Effects of Diet and Container Type on Growth
and Survival of Zebrafish Larvae ............................................................. 11
Characteristics of Artificial Microparticles Used in
Growth Experiments with Zebrafish Larvae ............................................ 17
Preparation of CLPWC ................................................................ 17
Preparation of GAB ..................................................................... 19
Retention efficiency of CLPWC ................................................. 22
Retention efficiency of GAB ....................................................... 23
Acceptability of CLPWC and GAB ............................................ 23
Digestibility of CLPWC and GAB .............................................. 24
Size of CLPWC Selected by Zebrafish Larvae ........................................ 25
Number of CLPWC per mg ......................................................... 26
Sinking rate of CLPWC ............................................................... 26
Feeding experiments with zebrafish larvae .................................. 28
Growth and Survival of Zebrafish Larvae Fed on CLPWC
or GAB as a Substitute for Brine Shrimp, Artemia sauna ...................... 32
RESULTS............................................................................................................... 34
Effect of Diet and Container Type on Growth
and Survival of Zebrafish Larvae ............................................................. 34
TABLE OF CONTENTS (Continued)
Page
Characteristics of Artificial Microparticles Used in
Growth Experiments with Zebrafish Larvae .............................................. 37
Retention efficiency of CLPWC ................................................. 37
Retention efficiency of GAB ....................................................... 38
Acceptability of CLPWC and GAB ............................................ 39
Acceptability of CLPWC and GAB by first feeding
larvae(ML, 3.82 mm) ............................................................... 39
Acceptability of CLPWC and GAB by 14 day-old
larvae (ML, 4.24 mm) .............................................................. 41
Acceptability of CLPWC and GAB by 20 day-old
larvae (MLS 5.2 mm) ................................................................. 43
Digestibility of CLPWC and GAB .............................................. 45
Size of CLPWC Selected by Zebrafish Larvae ........................................ 50
Number of CLPWC/mg ............................................................... 50
Sinking rate of CLPWC ............................................................... 50
Feeding experiments with zebrafish larvae .................................. 52
Growth and Survival of Zebrafish Larvae Fed on CLPWC
or GAB as a Substitute for Brine Shrimp, Arteinia sauna ...................... 59
Growth and survival of zebrafish larvae fed on CLPWC
as a substitute for brine shrimp, Artemia sauna .......................... 59
Growth and survival of zebrafish larvae fed on GAB
as a substitute for brine shrimp, Artemia sauna.......................... 62
DISCUSSION....................................................................................................... 66
Effects of Diet and Container Type on Growth
and Survival of Zebrafish Larvae ............................................................ 66
Characteristics of Artificial Microparticles Used in
Growth Experiments with Zebrafish Larvae ........................................... 67
Retention efficiency of CLPWC and GAB ................................. 67
Acceptability of CLPWC and GAB ............................................ 68
Digestibility of CLPWC and GAB .............................................. 69
Size of CLPWC Selected by Zebrafish Larvae ....................................... 70
Sinking rate of CLPWC ............................................................... 70
Feeding experiments with zebrafish larvae .................................. 70
TABLE OF CONTENTS (Continued)
Page
Growth and Survival of Zebrafish Larvae Fed on CLPWC
or GAB as a Substitute for Brine Shrimp, Artemia sauna ..................... 72
Growth and survival of zebrafish larvae fed on CLPWC
as a substitute for brine shrimp, Artemia sauna .......................... 72
Growth and survival of zebrafish larvae fed on GAB
as a substitute for brine shrimp, Arternia sauna .......................... 73
CONCLUSION ....................................................................................................... 74
BIBLIOGRAPHY ................................................................................................... 76
APPENDICES........................................................................................................ 81
LIST OF FIGURES
Figure
Page
1.
Imhoff conical shaped, transparent container used to
determine the effects of diet and container type on growth
and survival of zebrafish larvae ........................................................... 13
2.
Cylindro-conical, transparent "Coke" bottle used to
determine the effects of diet and container type on growth
and survival of zebrafish larvae ............................................................ 14
3.
Rectangular container used to determine the effects of diet
and container type on growth and survival of zebrafish larvae ............ 15
4.
Zebrafish larvae rearing system ............................................................ 18
5.
CLPWC containing Poly-R 478 used in feeding experiments .............. 20
6.
GAB containing Poly-R 478 used in feeding experiments ................... 21
7.
Final ML, of larval zebrafish at the end of 21 days .............................. 35
8.
Percent survival of larval zebrafish at the end of 21 days .................. 36
9.
Mean gut fullness and feeding incidences of first feeding
zebrafish larvae (5 day old; ML, 3.82 mm) fed on either MF,
CLPWCor GAB ................................................................................... 40
10.
Mean gut fullness and feeding incidences of zebrafish larvae
(12 day-old; ML, 4.24 mm) fed on either MF, CLPWC or GAB .........42
11.
Mean gut fullness and feeding incidences of zebrafish larvae
(19 day-old; ML, 5.22 mm) fed on either MF, CLPWC or GAB ......... 44
12.
Five day-old, first feeding zebrafish larvae (lateral view)
fed on CLPWC containing a red dye (Poly-R 478) .............................. 46
13.
Five day-old, first feeding zebrafish larvae (lateral view)
fed on GAB containing a red dye (Poly-R 478) .................................... 47
14.
Fecal strands collected from 15 day-old zebrafish larvae
(lateral view) fed on CLPWC containing a red dye (Poly-R 478)........48
15.
Fourteen day-old zebrafish larvae (lateral view) fed on GAB
containing a red dye (Poly-R 478) ........................................................ 49
LIST OF FIGURES (Continued)
Figure
Page
16.
Comparison of experimental data and predicted concentrations
showing the rate of loss of CLPWC during a period of 1 h,
under a flow rate of 80 ml/min. a) <20 jim, b) 2 1-45 jim,
c) 46-75 jim, d) 76-106 jim and e) 107-2 12 jim capsules ...................... 53
17.
Comparison of the predicted concentrations of different size
classes of CLPWC suspended in the water column, during 60 mm
with larvae of either ML, of 3.8 or 5.2 mm after accounting for the
effects of both sinking and ingestion by larvae ..................................... 57
18.
Comparison of mean lengths of zebrafish larvae fed on
Artemia nauplii substituted with different levels of CLPWC
atthe end of 8 days ............................................................................... 60
19.
Comparison of mean survival of zebrafish larvae fed on
Arternia nauplii substituted with different levels of CLPWC
atthe end of 8 days ............................................................................... 61
20.
Comparison of mean lengths of zebrafish larvae fed on
Artemia nauplii substituted with different levels of GAB
atthe end of 8 days ............................................................................... 64
21.
Comparison of mean survival of zebrafish larvae fed on
Artemia nauplii substituted with different levels of GAB
atthe end of 8 days ................................................................................ 65
LIST OF TABLES
Table
Page
1.
The weights of different size classes of CLPWC added to
Imhoff cones for determination of sinking rates to give a
total concentration of either 50 or 100 capsules/mI .................................... 27
2.
The weights of CLPWC of different size classes added to
each container at 15 minute intervals, during a period of 1 h,
in order to maintain the average concentration of each size class
at2 capsules /ml .......................................................................................... 29
3.
Mean concentrations and percent loss of Poly-R dye lost from
CLPWC suspended in either 10 ml dH2O or 10 ml 1M NaOH
after 2, 4, and 12 h of suspension ................................................................ 38
4.
Mean concentrations and percent loss of PoIy-R dye after
2,4, and 12 h suspension of GAB in dH2O ................................................ 38
5.
The mean numbers of CLPWC/mg prepared for this study (n=3) .............. 50
6.
Mean number of CLPWC/ml measured at 15 mm intervals
during a period of 60 mm under a flow rate of 40 mi/mm (n=3) ................ 51
7.
Mean number of CLPWC/ml measured at 15 mm intervals
during a period of 60 mm under a flow rate of 80 mI/mm (n=3)................ 51
8.
Size distribution and concentration of capsules in the foreguts
of 3.8 and 5.2 mm long zebrafish larvae fed on CLPWC for a
period of 60 mm at a flow rate of 80 mi/mm in Imhoff cones (n=l00) ...... 52
9.
Diet size selection of zebrafish larvae (MLS 3.8 mm) fed on
CLPWC during 60 mm with a flow rate of 80 mI/mm ................................ 58
10.
Diet size selection of zebrafish larvae (MLS 5.2 mm) fed on
CLPWC during 60 mm with a flow rate of 80 mi/mm ................................ 58
LIST OF APPENDIX TABLES
Table
Page
A. 1.
Multifactor ANOVA table for a comparison of the effect
of diet and container type on growth of zebrafish larvae ............................ 82
A. 2.
Tukey HSD comparison table for the effect diet and container
type on growth of zebrafish larvae after 8 days of feeding......................... 82
A. 3.
One way ANOVA table for differences among percent survival
rates of zebrafish larvae fed on either MF or MF+ART ............................. 83
B. 1.
Simple regression table for the absorbances of different
concentrations of Poly-R in distilled water................................................. 84
B. 2.
Simple regression table for the absorbances of different
concentrations of Poly-R in 1 M NaOH ...................................................... 84
C. 1.
One way ANOVA table for differences among gut fullness
values of first feeding zebrafish larvae (MLS 3.8 mm) fed on
MF, CLPWC or GAB as well as starved larvae ......................................... 85
C. 2.
Tukey HSD comparison table for effect of diet type
(MF, CLPWC, GAB or starved larvae) on gut fullness
values of first feeding
3.8 mm) zebrafish larvae ............................... 85
C. 3.
One way ANOVA table for differences among feeding incidences
of first feeding zebrafish larvae
3.8 mm) fed on MF,
CLPWCor GAB ......................................................................................... 86
C. 4.
Tukey HSD comparison table for effect of diet type
(MF, CLPWC or GAB) on feeding incidences of first
feeding (MLS 3.8 mm) zebrafish larvae ...................................................... 86
D. 1.
One way ANOVA table for differences among gut fullness
values of 12 day-old zebrafish larvae (MLS 4.2 mm) fed on
MF, CLPWC or GAB as well as starved larvae ......................................... 87
D. 2.
Tukey HSD comparison table for effect of diet type
(MF, CLPWC, GAB or starved larvae) on gut fullness
values of 12 day-old
4.2 mm) zebrafish larvae.................................. 87
LIST OF APPENDIX TABLES (Continued)
Table
Page
D. 3.
One way ANOVA table for differences among feeding incidences
of 12 day-old zebrafish larvae (MLS 4.2 mm) fed on MF,
CLPWCor GAB ......................................................................................... 88
D. 4.
Tukey HSD comparison table for effect of diet type
(MF, CLPWC or GAB) on feeding incidences of
12 day-old (MLS 4.2mm) zebrafish larvae .................................................. 88
E. 1.
One way ANOVA table for differences among gut fullness
values of 19 day-old zebrafish larvae (MLS 5.2 mm) fed on
MF, CLPWC or GAB as well as starved larvae ......................................... 89
E. 2.
Tukey HSD comparison table for effect of diet type
(MF, CLPWC, GAB or starved larvae) on gut fullness
values of 19 day-old
5.2 mm) zebrafish larvae.................................. 89
E. 3.
One way ANOVA table for differences among feeding incidences
of 19 day-old zebrafish larvae (MLS 5.2 mm) fed on MF,
CLPWCor GAB ......................................................................................... 90
E. 4.
Tukey HSD comparison table for effect of diet type
(MF, CLPWC or GAB) on feeding incidences of
19 day-old (MLS 5.2 mm) zebrafish larvae ................................................. 90
F. 1.
One way ANOVA table for differences among cones in the
growth of zebrafish larvae (17 day-old) pre-conditioned on a diet
of MF for 10 days followed by Arteinia nauplii for 2 days ........................ 91
F. 2.
One way ANOVA table for differences among cones in survival
of zebrafish larvae (17 day-old) pre-conditioned on a diet of
MF for 10 days followed by Arternia nauplii for 2 days ............................. 91
F. 3.
One way ANOVA table for differences among growth of
zebrafish larvae (25 day-old) fed on Artemia nauplii substituted
with different levels of MF encapsulated within CLPWC after
8 days of feeding ......................................................................................... 92
F. 4.
Tukey HSD comparison table for effect of substitution of
Artemia nauplii with different levels of CLPWC on growth
of zebrafish larvae after 8 days of feeding .................................................. 92
LIST OF APPENDIX TABLES (Continued)
Table
Page
F. 5.
One way ANOVA table for differences among survival of
zebrafish larvae (25 day-old) fed on Artemia nauplii substituted
with different levels of MF encapsulated within CLPWC after
8 days of feeding ......................................................................................... 93
F. 6.
Tukey HSD comparison table for effect of substitution of
Arremia nauplii with different levels of CLPWC on survival
of zebrafish larvae after 8 days of feeding .................................................. 93
G.1.
One way ANOVA table for differences among cones in the
growth of zebrafish larvae (17 day-old) pre-conditioned on a diet
of MF for 10 days followed by Artemia nauplii for 2 days ........................ 94
G. 2.
One way ANOVA table for differences among cones in survival
of zebrafish larvae (17 day-old) pre-conditioned on a diet of MF
for 10 days followed by Arteinia nauplii for 2 days ................................... 94
C. 3.
One way ANOVA table for differences among growth of
zebrafish larvae (25 day-old) fed on Artemia nauplii substituted
with different levels of MF encapsulated within GAB after
8 days of feeding ......................................................................................... 95
G. 4.
Tukey HSD comparison table for effect of substitution of
Artemia nauplii with different levels of GAB on growth
of zebrafish larvae after 8 days of feeding .................................................. 95
G. 5.
One way ANOVA table for differences among survival of
zebrafish larvae (25 day-old) fed on Arteinia nauplii
substituted with different levels of MF encapsulated
within GAB after 8 days of feeding ............................................................ 96
G. 6.
Tukey HSD comparison table for effect of substitution of
Artemia nauplii with different levels of GAB on survival
of zebrafish larvae after 8 days of feeding .................................................. 96
Growth and Survival of Zebrafish Larvae, Brachydanio
Fed on Two Types of Microparticulate Diets
rerio,
INTRODUCTION
Aspects of Larval Fish Development, Feeding and Digestion Physiology
Fish larvae undergo extensive morphological and physiological processes during
development including changes in respiration, swimming mode, energy metabolism, and
digestive system. (Segner
et al.,
1993). With respect to digestive system, fishes that
develop from short, post-hatch yolksac periods (termed altricial larvae), such as most
marine fish larvae, usually have difficulty utilizing artificial diets until they complete
metamorphosis. In contrast some fish species, such as salmonids, receive nutrients from a
yolk sac for a long period after hatching and develop a functional stomach after absorption
of the yolksac (termed precocial larvae). These latter species can successfully utilize
prepared diets from the time they first accept food.
The alimentary canal of precocial larvae is more developmentally advanced than that
of altricial larvae that hatch from small, positively buoyant eggs. At hatching, the
alimentary canal of altricial larvae is a straight tube, lying dorsal to the yolksac, which is
closed at the mouth and anus (Govoni et
al.,
1986). After the completion of yolk and oil
globule absorption, the undifferentiated tube becomes segmented by muscular valves onto a
buccopharynx, fore, mid, and hindgut (Govoni
et al.,
1986). The liver and pancreas are
formed at hatching and are functional by the end of yolk absorption (O'Connel, 1981).
With the exception of some precocial larvae, fish larvae lack both a morphological and
functional stomach but the foregut and midgut can expand and function in the storage of
food. The larval alimentary canal remains unchanged until the onset of metamorphosis and
the development of a functional stomach and pyloric caeca from the posterior foregut marks
the end of the larval period (Govoni etal., 1986) indicating the start of weaning.
'1
The sites of digestive enzyme function in fish include pancreas, gall bladder and
intestinal walls. Additional sources of enzymes come from live prey and bacteria. Enzyme
assays of the alimentary canal of fish larvae indicate that pepsin, trypsin, chymotrypsin and
amylase activities are apparent in several freshwater and marine fish larval species but the
activities of these enzymes are low at first feeding and increase during the larval period
before metamorphosis (Govoni et al., 1986). Intracellular digestion has been reported in
the larvae of a freshwater cottid, Cottus nozawae, by Watanabe (1984) who observed
pynocytotic absorption and intracellular digestion of macromolecular proteins by lysosomes
in the cells of the hindgut. After the start of exogenous feeding, in precocial fish larvae,
digestion takes place in an acidic environment mediated by pepsin. In altricial larvae,
without a functional stomach, digestion of ingested food takes place in the larval intestine
(niid and hindgut) where the pH remains alkaline and trypsin-type activity accounts for the
pioteolytic activity (Walford and Lam, 1993). Pepsin, a secretion of the gastric mucosa, is
not apparent until the gastric glands in the developing stomach become functional during
metamorphosis (Govoni etal., 1986). In seabass, Dicentrarchus labrax, larvae for
example, pepsin activity was detected for the first time on day 24 in the pancreatic segment
which included the stomach (Zambonino and Cahu, 1994). The mechanisms of protein
digestion and absorption changes from pynocytosis and intracellular digestion to
extracellular digestion and membrane transport with the development of gastric glands at
metamorphosis (Govoni etal., 1986). Compared to protein digestion, there is much less
information on lipid and carbohydrate digestion in fish larvae. Although lipids are broken
down to fatty acids and monoglycerides in the midgut lumen and there is evidence of
carbohydrate digestion, the mechanisms of lipid and carbohydrate digestion in fish larvae
are not known (Govoni etal., 1986). In altricial fish larvae, complete digestion of proteins,
lipids, carbohydrates, and other dietary components can not be achieved before the
development of a functional (secretory) stomach.
Several reasons have been offered for the poor ability of altricial larvae to utilize
prepared diets for growth. Some authors have proposed that the digestive system of larvae
is primitive and partly relies on the exogenous digestive enzymes from their live prey to aid
digestion (Dabrowski 1979; Walford etal., 1991; Holt, 1993). Lauf and Hoffer (1984)
estimated that in first feeding whitefish,Coregonus sp., larvae, exogenous proteases from
live prey represent up to 70-80% of the total proteolytic activity present within the digestive
tract. In addition to exogenous proteolytic activity, another possible contribution of live
prey is to induce an increase in endogenous trypsin secretion (Pedersen et al., 1987).
Although the breakdown of live food organisms may take place in the intestine of larvae
due to the combined effect of exogenous proteolytic enzymes and endogenous trypsin
secretion, extracellular digestion based on trypsin-type enzyme activity may not be
sufficient to achieve complete hydrolysis of proteins in the very short larval gut (Walford et
al., 1993). On the other hand, Govoni etal. (1986) suggested that the functional capacity
and the developmental status of the larval digestive system is sufficient to support growth
with live diets from first feeding onwards. In striped bass, Morone saxatalis, Baragi and
Lovell (1986) showed that at the time of first feeding, digestive enzymes are present at
sufficient concentrations to digest exogenous sources of nutrients. However, Weinhart and
Rosch (1991) concluded that one of the reasons for the lower growth rate of coregonid
larvae fed on dry diets was the result of low amount of dry diet eaten by the larvae
compared to larvae fed on Arteinia nauplii. A similar result has been reported by Kolkovski
et al. (1993) who reported a lower ingestion and assimilation of microdiet in gilthead
seabream,Sparus aurata, larvae. Recently, Cahu and Zambonino (1994) demonstrated that
seabass larvae were able to modulate their digestive enzyme activities in response to a
change in diet. Nevertheless, the same authors reported a lower secretion of pancreatic
enzymes in weaned larvae and supported the theory that the developmental sequence of
digestive functions of fish larvae can be advanced or delayed in response to diet
composition at weaning (Cahu etal., 1997).
ru
Fish larvae undergo a pattern of trophic ontogeny, changing diet with increasing
body size which reflects consequent changes in digestive development. Currently, there are
two major trends in larval fish feeding: 1) use of live food organisms such as rotifers,
Brachionus plicatilis, and crustaceans, Artemia sauna, and 2) use of formulated microdiets
(Kolkovski etal., 1993). Fry production of commercially important altricial marine larvae
such as seabass, gilthead seabream and turbot, Scophtalrnus maxiinus, and other temperate
marine larvae such as cod, Gadus morhua, and halibut, Hippo glosssus hippo glossus, is
still based on live food utilization for at least the first month of life (Person Le Ruyet et al.,
1993). Although rotifer and brine shrimp certainly hold many benefits as food for fish
larvae, commercial-scale production of both food organisms is costly due to the need for
space, energy, equipment and manpower. For example, Coves et al. (1991), calculated that
feeding the fish with live prey in intensive rearing conditions represents up to 79% of the
production cost of 45 day old seabass. Furthermore they are often unreliable in supply and
deficient in highly unsaturated fatty acids (HUFA). A number of recent studies have shown
that the HUFA, eicosapentaenoic acid (EPA; 20:5n-3) and docosahexaenoic acid (DHA;
22:6n-3) are essential dietary components for marine fish larvae (Webster and Lovell,
1990; Lemm and Lemarie, 1991). Marine fish larvae which are incapable of de novo
synthesis of essential fatty acids, depend on dietary sources for normal growth and
survival (Watanabe, 1993). Therefore, live food organisms need to be enriched prior
to being fed to fish larvae to improve their nutritional value and to meet the specific
nutritional requirements of fish larvae. Several enrichment preparations are now available
commercially in the form of oil-based concentrates, dried microcapsules and dry powders
which are relatively expensive and have a limited shelf-life because of their potential for
lipid oxidation during storage (Southgate, 1995).
Recent studies showed that altricial fish larvae can be successfully fed on artificial
diets as partial replacements for, or supplements to, live feeds (Jones etal., 1993). Larvae
of some freshwater species like whitefish, Core gonus lavaratus, common carp, Cyprinus
5
carpio, Ayu, Plecoglossus altivelus, and smailmouth bass, Micropterus dolomievi, have
been reared entirely on artificial diets (Jones etal., 1993); however, for marine fish larvae,
replacement of live food has not been accomplished. Up to 90% substitution has been
achieved for red bream, Pagrus major, and Japanese flounder, Para!ichlys o!ivaceus, with
10 day old larvae (Kanazawa et al., 1989). Tandler and Kolkovski (1991) showed that
gilthead seabream larvae can be reared successfully from first feeding with 50% live fed
substitution and they suggested that up to 80% substitution with microdiets is possible
without impairing growth. In some cases, co-feeding live and artificial diets can produce
growth and survival equal to that achieved with live feeds alone. This form of earlier
weaning is desirable because it allows reduced dependence on live feeds (Holt, 1993).
With the use of novel particle types for diet delivery, some of the difficulties in developing
artificial diets for fish larvae can be overcome (Langdon etal., 1985).
Development of Microparticulate Diets and Their Use for Fish Larvae
Microencapsulation has emerged as a means of nutrient delivery that may potentially
overcome problems associated with nutrient leaching and low water stability (Ozkizilcik
and Chu, in press). Various microencapsulation techniques have been developed for
delivering essential nutrients to crustacea, mollusks and fish larvae. For example, modified
nylon-protein microcapsules containing hemoglobin, (first described by Chang et al.,
1966) were used by Jones etal. (1974) and Levine et al. (1983) to study the nutritional
requirements of crustacean larvae. Calcium-alginate particles (Sommerville, 1962) were
adapted by Levine etal. (1983) for feeding to crustacean larvae. Gelatin-acacia walled
microcapsules were used by Langdon and Waldock (1981) in growth experiments with
Pasific oyster, Crassostrea gigas juveniles (Langdon et al., 1985). Microbound diets have
been used for the study of larval fish nutrition, such as red seabream and Japanese flounder
(Kanazawa etal., 1989). Tripalmitin-walled capsules have been successfully used to
deliver glycine to shrimp larvae by Villamar and Langdon (1993). Complex particles
consisting of alginate-bound gel particles with embedded lipid-walled microcapsules
(LWM) were described by Villamar and Langdon (1993) and used to show that shrimp
larvae could ingest and breakdown these complex particles. Recently, LWM encapsulated
within cross-linked protein walled capsules have been shown to be ingested and digested
by striped bass larvae (Ozkizilcik and Chu, in press).
Although much research has been done on microparticulate diets for larval fish,
a live food equivalent has not been achieved. The major problems encountered with
microparticulate diets are related to the intrinsic nature of the particles, such as texture, size,
digestibility and leaching of nutrients. For example, Lopez-Alvarado
et al.
(1994) reported
that carrageenan, alginate and zein microbound particles, as well as protein-walled
capsules. lost 60-90% of an amino acid mixture within 2 minutes after suspension in water.
Protein-walled and calcium-alginate microcapsules were shown to release low molecular
weight compounds rapidly (Langdon, 1983). Lipid wall capsules, on the other hand, are
capable of retaining water soluble compounds such as vitamins and minerals but are not
capable of delivering high nutrient loads to meet bulk nutritional needs (Langdon and
Siegfried, 1984).
In addition to the problems related to the intrinsic nature of microparticles,
difficulties related to the size of microparticulate diets have been reported. Walford et
al.
(1991) suggested that small particles can hardly be detected by fish larvae and large ones
can cause blockage of the digestive tract. It has also been shown that, although marine fish
larvae ingest microparticles, greater acceptance of living as opposed to inert prey has been
observed (Fernandez-Diaz et
al.,
1994). One of the objectives of this study was to
determine which size range of particles were ingested at the greatest rate by zebrafish,
Brachydanio rerio,
larvae. These data were then used in the design of microparticulate diets
for growth experiments with zebrafish larvae.
7
Availability of food to fish larvae is a crucial factor which has direct effect on larval
behavior, feeding success, growth and survival and it must be considered in the design of
the culture system as well as other conditions such as tank size, light intensity, flow rate,
aeration, temperature, and salinity. In order for microparticles to be digested, they must be
attractive and must be presented to the larvae under proper conditions (Langdon and Rust,
unpublished). Because sensory systems of early fish larvae are not fully developed,
environmental conditions which provide good sensory acuity are probably necessary to
maximize ingestion rates (Langdon and Rust, unpublished). The first goal of this study
was to test the effect of three different rearing units on the growth and survival of zebrafish
larvae fed on two different diets for 3 weeks. Using three different containers resulted in
various flow patterns affecting the distribution of food, behavior, growth and survival of
the larvae.
In the present study, a commercial larval food (Microfeast® Plus L- 10, Provesta
Corporation) was delivered to zebrafish larvae by encapsulation within two alternative
types of microparticles; cross-linked protein walled capsules (CLPWC) and alginate-gelatin
beads (GAB). Microfeast® (MF) is a formulated 5-10 tm, yeast-based (Torula sp.) diet for
fish and crustacean larvae. Preliminary feeding experiments with zebrafish larvae fed on
MF showed that this diet can be successfully used for the first feeding and promotes
growth and survival during 10 days after the start of exogenous feeding (Onal, pers. obs.).
The second goal of this study was to compare the leaching properties, acceptability, and
digestibility of CLPWC and gelatin-alginate beads GAB. After the comparison of different
culture units and the determination of the physical characteristics of CLPWC and GAB,
two experiments were carried out to determine the growth and survival of zebrafish larvae
fed on Arteinia sauna nauplii substituted with different levels of CLPWC and GAB.
The feeding experiments were carried out using zebrafish larvae as a model. This
species is a member of Cvprinidae and is native to the rivers of India and Bangladesh
(Molven et al., 1993). Zebrafish posses a number of experimental advantages for use in
feeding studies. They are available world-wide in pet stores and can easily be kept and bred
in the laboratory. The generation time is about 3 months and the adults are only 3-4 cm in
length. Zebrafish larvae are altricial, transparent and the gut contents can easily be observed
under a microscope until 30 days old, making them very suitable for feeding experiments.
In addition, zebrafish larvae readily accept a wide range of microparticulate diets (Rust,
1993) and can easily be cultured without high mortality rates under laboratory conditions.
Using zebrafish as a model in feeding studies can facilitate rapid research and development
of food particles and feeding conditions suitable for raising other species of altricial larvae.
MATERIALS AND METHODS
General Methods
Culture system for broodstock
Zebrafish eggs were obtained from broodstock maintained in the laboratory and
conditioned to spawn. Adult fish were fed manually 3-4 times a day on a mixture of freezedried tubifex worm (The Wardley Co.), flake food (Tetramin), and adult frozen
Artemia
sauna (Golden Gate Brine Shrimp; Kordon Division, Novalek, Inc.). Ten to fifteen fish,
with a female to male ratio of 1:2, were kept in six, 10 1 aquaria. A flow-through system
was used to maintain optimum culture conditions for the broodstock. Tap water was first
passed through a carbon filter to remove chlorine then pumped through two cartridge filters
(100 tm and 20 m) and into a rectangular head tank (500 1 in volume). Water temperature
was maintained at 28 °C by two thermostat-controlled heaters (1800 W and 1000 W;
Cleveland Process Corporation) immersed in the head tank. Strong aeration was provided
to the head tank to ensure even heat distribution and oxygenation. The water was then
passed over a U.V lamp (80W; Rainbow Plastics Filter Division) before being delivered
into the broodstock tanks. In order to maintain good water quality, uneaten food and faeces
were removed by siphoning every two days. During the study (9 months), mortality of
broodstock in the flow-through system was less than 5%. Photoperiod was manually
controlled by fluorescent lights (12-14L: l0-12D), the dark period was eliminated several
days before spawning to prevent uncontrolled breeding of the fish.
Breeding
On the day before the eggs were needed, 1-2 h before the end of the light period,
the fish were fed ad
libitum
with frozen brine shrimp and the bottom of each aquarium was
siphoned and covered with plastic screens (mesh size of 3 mm). In order to keep the plastic
screens in position, small stones were glued to the edges and center of both surfaces. By
10
covering the bottom of the tank with plastic mesh, breeding fish were unable to eat the
freshly laid eggs because the eggs sank and passed through the mesh. Early in the morning
of the next day, lights were turned on, and light intensity increased gradually over 2-3 h.
During this period spawning groups were observed by their characteristic fast swimming
and chasing activity which resulted in release and fertilization of eggs. After 2-3 h, eggs
were collected by siphoning the bottom of the tanks. Faeces and other residues were
removed from the eggs with a pipette. To help prevent mold, the eggs were washed for 2
minutes with a 1 ppt solution of commercial bleach (5.25% NaC1O by weight). The eggs
were then washed with freshwater, counted and placed into hatching containers made from
plastic baskets that were covered with Nitex screen (200 jim mesh). A gentle water flow
at 28 °C was provided and the dead eggs were removed everyday to prevent disease
outbreaks. At 28 °C, zebrafish eggs hatched 2 days after fertilization and the larvae started
swimming freely on day 4, spending most of the time resting on the sides of the containers.
With this hatching method, egg mortality was about 5% and 4-5 thousand larvae were
obtained in each trial. All experiments were started with the same age-class of larvae
(5 days old) that were free swimming and had consumed most of the yolk sac.
Larval measurements
In this study, larval age was expressed as days after hatching. Before the start of
each experiment, all the larvae (5 days old) from one hatch were pooled together and
30 larvae were sampled to determine the average initial length of the population. In all
experiments, larval mean standard length (ML) was used as an indicator of growth.
Standard length (notochord length) from the tip of the mouth to the tip of the notochord
was measured with a computer-aided image analysis system (Macintosh Quadra 660 AV;
NIH Image 61). Percent survival rates were calculated using the following formula:
(Final # of surviving larvae! initial # of larvae) x 100
(1)
11
In the present study, the larvae used in the growth and survival experiments as
well as the digestibility experiments were anaesthetized (10 mg/i Tricaine; Finquel®,
Argent Chemical Laboratories), placed on a slide and measured alive under a dissection
microscope with 12X and 25X magnification depending on the size of the larvae. The
larvae in acceptability and diet size selection experiments were fixed with 70% alcohol
without anaesthetizing. Preliminary experiments showed that after 12 h in 70% alcohol
there was no shrinkage in the ML of larvae.
Water quality
Throughout the growth and survival studies, temperature, oxygen, pH and
ammonia measurements were taken. Temperature was measured with a thermometer
everyday and every other day the following measurements were taken: oxygen (YSI
oxygen meter; Model 51 B in the first experiment, and Strathkelvin Instruments; Model 781
in other experiments), pH (Corning Medical; pH meter 125) and ammonia (Hatch
Ammonia Kit).
Effect of Diet and Container Type on the Growth and Survival of Zebrafish
Larvae
A culture system was developed to compare the growth and survival of zebrafish
larvae fed on either Microfeast® (MF) or Microfeast®+brine shrimp, Artemia
sauna,
(MF+ART), in three different types of containers and flow configurations. The tested
containers were:
a. One liter, transparent Imhoff settlement cones (10.5 cm W x 45 cm D; Wheaton
brand; Figure 1),
b. Two liter, inverted, transparent cyiindro-conical "Coke" bottles (10.5 cm W x 25
cm D; Figure 2),
12
c. Three liter, rectangular, plastic containers (25 cm L x 14 cm W x 10 cm D;
Rubbermaid brand; Figure 3).
Both cones and cylindro-conical bottles were modified so that culture water entered
the containers from the bottom and left from the side, near the top, creating an upwelling
current. In the rectangular containers, water entered from the opposite side, close to the
bottom, and left from the other side, close to the top, resulting in a different flow pattern
compared to that of cones and cylindro-conical containers. This particular flow pattern was
the major difference between the containers. The outlets were hand drilled and their heights
adjusted to give total culture volumes of 1.2 1 in cones, 1.8 1 in cylindro-conical bottles and
2.4 1 in rectangular tanks. For each container, screens (200 tm mesh) were attached to the
inlet and outlet of each container in order to prevent the larvae from escaping. In order to
keep the containers in position, a stand was fabricated using 1/2 PVC piping. The water
entered the system by gravity and the flow rate was controlled by plastic valves (Aquatic
Ecosystems; part no.VPL1).
Throughout the duration of the experiments, the laboratory was kept under
controlled temperature and photoperiod conditions of 28 °C and 14L: 1OD, respectively.
Six, 15 W fluorescent lights provided a mean light intensity of 10 p. E s' m2 (=6.022 x
io'
photons s' m2; Li-Cor light meter; Model no: Li-l85B). No aeration was used in the
culture vessels to generate water currents or dispense food. The flow rates were 40 ml,
53.3 ml, and 80 mi/mm in cones, cylindro-conical bottles and rectangular containers,
respectively, so that in each container, two turnovers of culture volume per h were
maintained throughout the duration of the experiment.
Fourteen day-old larvae were randomly assigned to each treatment at a stocking
density of 40 larvae/i. Two diets with three replicates per treatment were tested with each
container type for a period of 7 days. The number of replicates were determined using the
following formula (Steel and Torrie, 1980):
13
I
10.5cm
E
Figure 1: Imhoff conical shaped, transparent containers used to determine the effects of
diet and container type on growth and survival of zebra fish larvae. Total volume of the
container was 1.2 1 with a flow rate of 40 mi/mm and the stocking density was 40
larvae/i. A) Imhoff cones, B) outlet, C) outlet filter, D) inlet filter.
14
I
10.5cm
I
E
U,
Figure 2: Cylindro-conical, transparent "Coke" bottle used to determine the effects of diet
and container type on growth and survival of zebrafish larvae. Total volume of the
container was 1.8 1 with a flow rate of 53.3 mi/mm and the stocking density was 40
larvae/I. A) Coke botle, B) outlet, C) outlet filter, D) inlet filter.
15
25 cm
A
Figure 3: Rectangular container used to compare the effects of diet and container type on
growth and survival of zebrafish larvae. Total volume of the container was 2.4 1 with a
flow rate of 80 ml/min and the stocking density was 40 larvae/l. A) Rectangular
container, B) inlet, C) outlet filter, D) outlet.
r
2(Zw2+Z$)2
(a
(2)
where r = number of replicates, Za/2 = 1025 = 1.96, Z = Z.io = 1.28, a2 = estimate of
a from previous experiments and 8 = the size of difference to be detected (-2.5%).
In one dietary treatment, larvae were fed on a ration of 1.04 mg/fish/day MF every
hour for a period of 10 h. In the second treatment, the larvae were fed on the same ration
for the first 14 days and then a combination of 1.04 mg/fish/day of MF and an increasing
concentration of Artemia nauplii starting with 35 Jig/fish/day (20 nauplii/fish/day) on day
14 and ending with 350 tg/fish/day on day 21(200 nauplii/fishlday) over a period of 10 h
each day. Every hour 134.78 mg of MF was suspended in 270 ml of tap water (28 °C) for
2-3 minutes and 10, 15 and 20 ml of suspensions were added to each cone, cylindroconical bottle and rectangular container, respectively. Arternia nauplii cysts (U.S.E.P.A.,
Reference Artemia Cysts Ill) were incubated for 24 h in aerated sea water (38 ppt) at 28 °C.
The nauplii were rinsed with freshwater and kept in 11, 30 ppt, 3-5 °C, aerated sea water.
The mean concentration of Artemia nauplii was determined by counting the number of
nauplii in I ml glass pipette (in triplicate) and nauplii were added to each container every
hour in given amounts.
At the end of 21 days, thirty larvae were sampled from each container and the
effects of diet and container type were evaluated by statistically comparing the mean growth
and survival of zebrafish larvae among treatments. Residual plots of the data were analyzed
to assure whether assumptions of ANOVA were met or not. All treatments were first
compared by Multifactor ANOVA at a significance level of p<O.O5. The treatments were
then compared by one way ANOVA followed by Tukey's Honest Significant Difference
(HSD) multiple range test in order to find which means were significantly (p<O.O5)
different from others.
17
It was found that using cones resulted in higher larval growth rates compared to
those with cylindro-conical bottles and rectangular containers (see results). Therefore, the
capacity of the larval system was doubled (a total of 24 cones) using Imhoff cones as the
rearing unit for zebrafish larvae (Figure 4). In this study, the remaining experiments with
microparticulate diets were carried out using Imhoff, conical shaped, one liter cones.
Characteristics of Artificial Microparticles Used in Growth Experiments
with Zebrafish Larvae
The results obtained from previous experiments showed that MF promoted growth
and survival of zebrafish larvae during the first 10-12 days of life; however, 8 days after
the larvae started feeding on Artemia nauplii, growth of zebrafish larvae fed on a mixture
of MF+Artemia nauplii was significantly greater than that of larvae fed on MF alone. Two
reasons could have contributed to the poor growth of larvae on MF alone: a) nutritional
composition of MF diet was not sufficient for optimal growth and survival of 15-20 day
old zebrafish larvae or b) particle size of MF was too small to meet the metabolic cost for
search and capture of food. With respect to this last point, MF was encapsulated within
different types of microparticulate diets resulting in larger sized particles. Two types of
microparticles were used; cross-linked protein walled capsules (CLPWC) or gelatinalginate beads (GAB). A comprehensive evaluation of the microparticulate diets were
carried out before they were fed to the larvae including determination of their retention
efficiencies for a water soluble dye and determination of their acceptability and digestibility
by zebrafish larvae.
Preparation of CLPWC
MF was encapsulated using a modified method described by Langdon (1989).
Five grams of egg albumin were dissolved in 15 ml of I M NaHCO3 to give a final
concentration of 33% w/v albumin. One gram of MF was added to the egg albumin
18
Figure 4: Zebrafish larvae rearing system. This flow-through system was designed for
feeding studies with microparticulate diets. Water entered the bottom of containers by
gravity, creating an upwelling current of 4.8 1/mm. Arrows indicate direction of flow. A)
Stand made from 1/2" Pvc pipes and an acrylic sheet, B) Imhoff settling cones (1.2 tin
volume), C) plastic valves, D) standard aquarium hose and E) drain pipes.
19
solution and sonicated for 15 sec. All the mixture was emulsified in 100 ml of 2% soy
lecithin dissolved in cyciohexane. The emulsion was then poured into a 300 ml beaker
and agitated on a magnetic stirrer. Fifty ml of 2% soy lecithin dissolved in chloroform
was then added to the stirred emulsion followed by I ml of cross-linking agent (sebacoyl
trichloride). After addition of the cross-linking agent, the aqueous emulsion was stirred for
15 minutes at room temperature to ensure that wall formation was complete. The capsule
suspension was then added to 250 ml of cyclohexane and the capsules allowed to settle out
of suspension for 6 hours. The settled capsules were then washed by 4 series of dilutions
of ethanol (100%, 75%, 50%, and 25%, respectively) and finally distilled water. The
capsule slurry was then frozen by swirling the sealed tube containing the capsules in a
bath of ethanol at -80 °C and freeze-dried. Freeze-dried capsules then sieved through 20,
45, 75, 106, 212 and 425 tm mesh sieves and the different size fractions were stored
separately (Figure 5).
Preparation of GAB
Previous experiments with zebrafish larvae suggested poor digestion of alginate
beads. Therefore, another approach was tested by making beads out of a mixture of gelatin
and alginate because gelatin would likely be digested easier by the larvae. GAB containing
MF were produced by the method developed by Langdon and Jacobsen (unpublished).
Three hundred mis of 2% w/w sodium aiginate (Colloid Technologies Center; TIC-GUMS,
Inc.) mixed with 200 mIs of 2% w/w gelatin (General Foods Co.) and 5% w/w MF were
sprayed into a collection chamber containing two liters of 20% w/v chilled (0-5 °C)
CaC12.
The gelatin-alginate mixture needed to be maintained at a temperature above 40 °C, so the
nozzle system of the spray device (l/4JBCJ; Spraying Systems Co.) was jacketed and 40
°C water was circulated through the housing. To optimize the delivery of gelatin-alginate
solution to the nozzle system, a pressurized chamber was used. The chamber was made
from 3" PVC tubing with a flat cap on the bottom and threaded cap on the top. In order to
20
0
C
.
0
0
0
.
C
S
.
iI
Figure 5. CLPWC containing Poly-R 478 used in feeding experiments. Magnification,
50X objective. Scale
=80 pm.
21
r
4O
S
Figure 6: GAB containing Poly-R 478 used in feeding experiments. Magnification, 50X
objective. Scale
= 80 JIm.
22
ensure that the MF was well mixed with the gelatin/alginate, the mixture was sonicated in a
specially constructed sonication chamber just before the mixture was passed through the
nozzle system. The gelatin-alginate mixture was forced through a 3600 spray nozzle where
it was atomized with nitrogen under pressure (90 psi). When the atomized droplets came
into contact with the chilled CaCl2, they formed beads ranging from 20 to 1000 tm in
diameter. Beads were rinsed with distilled water and poured through 425, 212, 106, 75,
45 and 20 tm mesh sieves. The different size fractions were then freeze-dried and stored
for use in feeding experiments (Figure 6).
Retention efficiency of CLPWC
CLPWC were prepared by the method described above and contained 3% w/v of
a high molecular weight, non-toxic dye (Poly-R 478; Sigma Chemical Co.). Poly-R was
used because it had a molecular weight of 50,000 to 100,000 Daltons (Buchal, 1994) and
would be retained by CLPWC with high efficiency. Retention efficiency was expressed as
the percentage of encapsulated core retained after suspending capsules in distilled water for
a known period of time. Retention efficiency was determined by comparing the amount of
leached dye in distilled water with the initial amount of encapsulated core material. The
amount of encapsulated core was determined by dissolving a known weight of capsules
in 1 M NaOH and measuring the amount of Poiy-R using a spectrophotometer (Beckman
DU-6 Spectro-photometer). Absorbance readings were converted to Poly-R concentrations
(mg/mi) using a simple regression equation derived from standard curves.
Maximum absorbance of Poly-R was found to be at 520 nm both in distilled water
and 1 M NaOH. Triplicate samples of 20 mg capsule were suspended in either 10 ml
distilled water or 10 ml 1M NaOH contained in sealed 20 ml glass tube. The absorbances
of the leached dye in 1 M NaOH and distilled water were then measured after 2,4 and 12
hours of suspension. Each sample was centrifuged (Beckman TJ-6 Centrifuge) at 5000
23
rpm for 5 mm before absorbances were measured in the supernatant solution at each sample
time. The amounts of encapsulated core released in NaOH and leached dye were compared
in order to estimate retention efficiency.
Retention efficiency of GAB
GAB containing 3% w/v of Poly-R were prepared as described above. Retention
efficiency of capsules was expressed as the percentage of the estimated encapsulated core
retained after a given suspension time. The initial amount of encapsulated core was based
on the assumption that all the core material was initially encapsulated within the wall
material after preparation of capsules. An estimate of the encapsulated core material was
necessary for GAB because it was not possible to dissolve the beads and extract the
core material (Poly-R), even though a variety of extraction solutions were tested such
as I N HC1, I N HCI+10% EDTA, I M NaOH and a carbohydrase solution containing B
glucuronidase and sulfatase (Abalone Acetone Powder; Sigma Chemical Co.).
In order to determine the retention efficiency of Poly-R by GAB, 10 mg of GAB
were suspended in 20 ml of distilled water and samples were taken after 2, 4 and 12 h
of suspension. Each sample was centrifuged at 5000 rpm for 5 mm (Beckman TJ-6
Centrifuge) and the absorbances of 3 ml samples of the supernatant were measured at
520 nm to determine the amounts of Poly-R lost from the suspended GAB. The losses
were then compared with the estimated initial encapsulated core concentration of Poly-R
in order to determine the percentage of the initial encapsulated Poly-R that was retained
after 2, 4 and 12 h of suspension.
Acceptability of CLPWC and GAB
Acceptability of microparticulate diets was expressed both in terms of both gut
fullness and feeding incidence of larvae fed on CLPWC or GAB in a given period of time.
The gut fullness of larvae fed on CLPWC or GAB was determined by measuring the area
24
of material contained within the foreguts of larvae (measured as side-view, cross sectional
area) using a computer aided image analysis system (Macintosh Quadra 660 AV; NIH
Image 61). Feeding incidence was determined by counting the proportion of larvae with
diet particles in their guts as observed through the larvae's transparent body tissues after a
given period of time. In order to determine the acceptability of microparticulate diets, larvae
were pre-starved over night to clear their guts of ingested material and then fed on CLPWC
or GAB for a pre-determined period of time. Control groups included starved larvae, and
acceptance of CLPWC or GAB by larvae was compared with that of larvae fed on MF. MF
was used as a control diet because preliminary experiments showed that acceptance of MF
by both first feeding larvae and larvae of older age (18-20 day old) was high. A similar
method of measuring particle acceptability has been successfully used by Dr. M. B. Rust
in experiments with goldfish, walleye and zebrafish.
Acceptability experiments were carried out at 28 °C, with a flow rate 80 mi/mm in
linhoff cones for 5, 13 and 20 day old zebrafish larvae with ML, of 3.7, 4.2 and 5.2 mm,
respectively. Preliminary experiments showed that larvae at first feeding (5 day old) filled
their guts in 2 h whereas larvae of greater age (13 day old; ML, 5.4 mm) filled their guts in
I h; therefore, larvae with a ML, of 3.7 mm were sampled 2 h after first feeding, 4.2 and
5.2 mm larvae were sampled after 1 h. The experiments were carried out in triplicate,
with 100 larvae added to each cone. After the feeding period, the fish were killed in 70%
alcohol, put on a slide and immediately video-taped. The mean gut fullness and feeding
incidences of zebrafish larvae fed on CLPWC, GAB or MF were determined by sampling
30 and 100 fish, respectively, from each container. The acceptabilities of the diets were
compared by ANOVA and Tukey's HSD multiple range tests (p<0.05).
Digestibility of CLPWC and GAB
The digestibilities of CLPWC and GAB were determined by observing the
alimentary canals of zebrafish larvae fed on these two microparticle types. For this
25
purpose, larvae were fed for several hours on either CLPWC or GAB containing an
aqueous solution of non-toxic, red polymeric dye (Poly-R 478; Sigma Chemical Co.).
The larvae were then sampled, placed on a slide and observed under a microscope (Zeiss
Reflected Light Microscope). Finally, the larva's digestive system was photographed using
a photo-micrographic camera (Zeiss MC 63) attached to a microscope, with a tungsten film
(Kodakrome T64 and T160) under 20x and 40x magnification. If the walls of the capsules
were ruptured, released dye was observed in the digestive systems of the larvae. In
addition, change in the shape of the microparticles in the digestive tract of the fish larvae
indicated breakdown of the microparticles. A similar experimental approach has been
successfully used by Villamar and Langdon (1983) in determining the digestibility of lipid
walled capsules by shrimp larvae.
Size of CLPWC Selected by Zebrafish Larvae
In order to optimize the effect of size of particles on the growth of zebrafish larvae,
a series of experiments was carried out to determine the size of diet selected by larvae in
two different length classes: first feeding larvae with a ML, of 3.8 mm and 13 day old
larvae with a ML, of 5.2 mm. Experiments on diet size selection were carried out with
CLPWC because these capsules had a rigid, spherical shape that could be easily seen and
measured in the stomach. On the other hand, GAB were irregular in shape and preliminary
experiments showed that they changed shape in the guts of larvae, making it difficult to
determine bead size.
Throughout the duration of the feeding experiments, equal numbers of five different
size classes of CLPWC were maintained in the water column. Maintaining equal numbers
of CLPWC throughout the duration of the experiment was only possible after determination
of the mean sinking rate of each size class of CLPWC.
Number of CLPWC per mg
The number of capsules of each size class per mg of capsule preparation was
calculated by counting the number of capsules in each size class in a given weight of
capsules. Capsules in size classes above 45 jim were counted in a dry powder preparation
under a dissecting microscope with 1 2X magnification. For this purpose 0.5 mg of freezedried CLPWC were examined for each size class. Capsules smaller than 45 jim were first
suspended in distilled water and a known volume of the suspension was sampled. For
this purpose, 0.1 g of freeze-dried CLPWC was suspended in 11 of distilled water and a
sample of 100 jil was taken by pipette. The sample was then placed on a slide and covered
with a cover slip. The capsules were counted under a microscope under 20x magnification
and the mean number of capsules in each size class per mg of capsule was determined.
Although the capsules were counted immediately after hydration, a small error in
determining the diameter of capsules due to their swelling in distilled water was possible;
however, change in diameter of capsules was determined to be <5% after 15 minutes of
suspension in distilled water and was considered negligible. Triplicate samples were taken
for each size class measurement.
Sinking Rate of CLPWC
A series of experiments (in triplicate) was carried out to determine the sinking rate
of CLPWC in Imhoff cones (total volume 1.2 1) with two different flow rates (40 and 80
ml/min). One ml samples of culture water at 15 minute intervals were taken and changes
in number of capsules/ml were determined over a period of 60 mm for each size class.
Samples were taken 15 cm below the surface where the larvae gathered. During the
experiments, a 10 jim Nitex screen attached to the outlet of each container prevented the
loss of capsules larger than 10 jim due to water exchange. Table 1 summarizes the weights
of each size class of CLPWC that were added to Imhoff cones to give initial concentrations
either 50 or 100 capsules/ml for each size range.
27
A comparison of changes in the concentrations of CLPWC with two different flow
rates showed that the suspension of capsules, particularly for sizes larger than 75 tm, was
increased when flow rate was increased from 40 to 80 ml/min (see results). Therefore,
sinking rates of CLPWC were only determined for the flow rate of 80 mI/mm, and size
selection experiments were also carried out with this flow rate.
Table 1: The weights of different size classes of CLPWC added to Imhoff cones for
determination of sinking
rates to give a total concentration of either 50 or 100 capsules/nil
**
100 capsules/mI, 50 capsules/ml). Flow rate through cones was 40 or 80 mI/mm.
Size class (lim)
Weight (mg) added per cone
(V7=1.2
<20
0.510*
21-45
1.766
46-75
7.807"
76-106
19.486**
107-212
149.253
I)
The general equation used to describe the exponential decrease in the concentration
of CLPWC during a period of 60 mm and with a flow rate of 80 mI/mm was:
x,(t) = C,e"
where x,
(t)
= number of capsules of a given size class at time t, in a volume of 1.2 1,
C. = initial concentration of CLPWC, e = 2.7818, a = rate at which particles were lost
from suspension,
t
= time in minutes,
(3)
Equation (3) was re-arranged so that the sinking rate of CLPWC during 60 mm
could be calculated from the standardized initial concentration of 10 capsules/mI for each
size class:
x(t) = 10e
(4)
a, the rate at which particles are lost from suspension, was determined for each size
class, by the rearrangement of equation (3) as follows:
logC0logC
(5)
where C'o=initial concentration, Ct= concentration at time t.
A fitted mean 'a' value was calculated for each size class by using experimental
data for samples taken at 15 minutes intervals, over a period of 60 mm with a flow rate of
80 ml/min. These fitted mean 'a' values for each size class were then used in the analysis
of particle size preferences of zebrafish larvae.
Feeding experiments with zebrafish larvae
A final set of experiments was carried out to determine the size of capsules
zebrafish larvae preferred to ingest when they were either 3.8 mm (5 days old; first
feeding larvae) or 5.2 mm in length (10-12 day-old larvae feeding on Artemia nauplii).
A total of 300 fish were placed in three cones (100 larvae/cone) and the flow rate was
adjusted to 80 mI/mm
(4
turnovers/h). The experiments were designed so that an average
of 2 capsules/mi of each size class were maintained in the water column by adding a pre-
determined weight of capsules from each size class every 15 minutes over a period of 1 h.
Due to differences in the rate of loss of each size class, different amounts of capsules
of each size class were added to each container in order to maintain equal average
concentrations of each size class over a period of 1 h (Table 2).
Table 2: The weights of CLPWC of different size classes added into each container at 15
minute intervals, during a period of 1 h, in order to maintain the average concentration of
each size class at 2 capsules /ml.
Size class (jtm)
Weight of each size class (mg)
added to lmhoff cones (VT=l.2
<20
0.001
21-45
0.004
46-75
0.106
76-106
0.530
107-212
4.179
I)
After a period of 1 h, all the larvae in the cones were sampled and preserved in
70% ethanol. Preliminary experiments with CLPWC showed that after 12 h suspension in
distilled water and 70% ethanol, the increase in diameter of the capsules was 5% compared
to the initial diameter. Final diameters of CLPWC were corrected for this effect. The larvae
were placed on a slide, their guts were opened with a needle and the gut contents were
forced out. The stomach samples of a total of 90 fish (30 fish per cone) were then
observed under a microscope and video-taped. The sizes of the ingested capsules were
then measured by a computer-aided image analysis system and feeding rate (the mean
number of capsules per gut examined) throughout the experiment was determined. The
particle size preference of zebrafish larvae as determined by accounting for different
sinking rates of capsules of different size and the size composition of ingested capsules.
30
A mathematical model to determine the particle size preference of zebrafish larvae was
developed based on the following assumptions:
a) the reduction in the concentration of particles was due to ingestion by the larvae
and to gravitational settling,
b) the larvae's ingestion rate was constant throughout the experiment,
c) the larvae were sampled before CLPWC were digested or egested,
d) experimental conditions were the same throughout the experiment.
The following equation was used to determine the particle size preference of
zebrafish larvae:
dC,(t)
dt
(6)
where dC1(t)/dt is the rate of decrease in concentration
at time t; a is the rate at which
particles were lost from suspension for a given size class; C, is the concentration of a given
size of capsule in the container; and Z is a constant obtained by the mean feeding rate/60
mm.
Solving the above equation gives:
C,(t)
where
C1
fl1e' --sa,
(7)
is the concentration of a given size class at time t andj3, is the initial concentration
for a given size class.
For t=0 to t= 15- (time before the first addition of capsules), the equation becomes:
31
C1(t) =
[c(o)
+
-
(8)
a1]
For t= i5 (time after the first addition of capsules) to t=30 (time before the second
addition of capsules),
C1(t) =
[ci(i5)+1e"
Cr1]
(9)
[
where C1(15)=c, (i5)+ö and 5 is the amount added every 15 minutes for a given size
class.
As shown in equations
(8)
and (9), the average concentration of capsules in each
size class in the water column was calculated for a period of 60 mm by accounting for
the rate of loss of capsules, ingestion by the larvae and addition of capsules from each
size class to compensate for sinking at 15 mm intervals. In order to maintain an average
concentration of 2 capsules/mi for each size class throughout the duration of the
experiment, a pre-determined weight of capsules for each size class (see Table 2) was
added to containers at 15 mm intervals (3 times) to compensate for loss due to sinking.
The selection coefficients for zebrafish larvae fed on the five different size classes
were then determined by the following equation:
Selection Coefficient =
Mean # of capsules of a given size class in the gut
Mean # of capsules of a given size in the culture volume
(10)
The coefficients obtained for each size class of CLPWC from equation (10) were
then normalized using following equation:
32
% Relative Selection =
Selection coefficient for a size class of capsule
x 100
selection coefficents for all size classes
(11)
The % relative selection coefficients for the various size classes of capsules were
then compared to each other in order to determine which size class was preferred by
zebrafish larvae either 3.8 or 5.2 mm in length.
Growth and Survival of Zebrafish Larvae fed on CLPWC or GAB as a
Substitute for Brine Shrimp, Artemia sauna
Two different experiments were carried out to compare the growth and survival
of zebrafish larvae fed on brine shrimp, Artemia sauna, substituted with different levels
of MF encapsulated within CLPWC or GAB. The initial stocking density was 100
larvae/container. The flow rate was maintained at 80 ml/min in order to enhance
niicroparticle suspension. Throughout the duration of the experiments, larvae were
kept under controlled temperature and photoperiod conditions of 28 °C and 14L: 1OD.
Six, 15 W fluorescent lights provided a mean light intensity of 10 jt E s m2 (=6.022 x
1017
photons s' m2).
At the beginning of each experiment, larvae in each group were pre-conditioned
on a diet of 100% MF (50
mg/day; 5 mg/h over 10 hours) for a period of 10 days followed
by a transition period of 2 days of feeding on Artemia sauna nauplii (10 mg/day). After
this pre-conditioning period, growth was determined by sampling 20 fish from each
container. Percent survival rates were determined by using equation (1) (page 10). All
the containers were then cleaned and during the cleaning process (1 day), the larvae were
starved. Treatments were then randomly assigned to each container (80 larvae/ container)
with three replicates per treatment. After the treatments were assigned to each container, the
growth and survival of larvae in the different cones were compared in order to determine
if there were any differences among treatments before starting to feed the larvae on different
33
levels of microparticles and Artemia nauplii (ANOVA; p<O.O5). In each experiment a total
of 8 treatments were tested:
Treatment 1: Starved,
Treatment 2: 100% MF (50 mg/day),
Treatment 3: 100% CLPWC or GAB (50 mg/day),
Treatment 4: 100% Arteinia nauplii (25 mg/day),
Treatment 5: 80% CLPWC or GAB+ 20% Artemia nauplii,
Treatment 6: 60% CLPWC or GAB+ 40% Artemia nauplii,
Treatment 7: 40% CLPWC or GAB+ 60% Artemia nauplii,
Treatment 8: 20% CLPWC or GAB+ 40% Artemia nauplii,
Starting on day 14, the larvae were fed daily (every hour for a period of 10 h)
on a mixture of different levels of CLPWC or GAB and Arte,nia nauplii as shown above.
Microparticulate diets were suspended in 450 ml of tap water (28 °C) for 2-3 minutes
before feeding the larvae every hour. The size spectrum of microparticulate diets given
to larvae consisted of 0.7% <20 im,27.2% 21-45 JIm, 61% 46-75 JIm, 10.2% 76-106
jim and 0.9% 107-212 jim size capsules on a weight basis.
Arternia nauplii cysts (U.E.S.P.A, Reference Artemia Cysts III) were incubated
for 24 h in aerated sea water (38 ppt) at 28 °C. The nauplii were rinsed with freshwater
and kept in a 11 graduated cylinder in 30 ppt, 3-5 °C, aerated, ice-chilled sea water. The
daily total number of Arternia nauplii added to each cone was determined by dividing
the total daily amount (25 mg/day) to individual dry weight of Artemia nauplii (1.78 Jig)
as reported for this brand (Crewell, 1992). The concentration of Artemia nauplii was
determined by counting the number of larvae in a I ml glass pipette.
On day 22, the experiment was terminated, 30 larvae were sampled from each
container and the growth and survival of zebrafish larvae in each treatment were compared
(ANOVA and Tukey's HSD multiple range procedure; p<O.O5).
34
RESULTS
Effect of Diet and Container Type on the Growth and Survival of Zebrafish
Larvae
There was a significant effect of diet (Multifactor ANOVA; p<O.00I; see Appendix
A, Table 1) on the growth of zebrafish larvae. At the end of the experiment, the mean
length of larvae from treatment MF was found to be 5.15 mm, the larvae from treatment
MF+ART had a mean length of 6.96 mm. In addition, there was a significant effect of the
shape of the containers on the growth of zebrafish larvae (Multifactor ANOVA; p=O.009).
The growth of the zebrafish larvae in both dietary treatments was highest in cones with a
final ML of 7.37 mm in treatment MF+ART and 5.56 mm in treatment MF (Figure 7).
Although interaction effect of diet and container was not significant (Multifactor ANOVA;
p= 0.1257), further analysis of the data showed that larvae fed on a mixture of MF and
Arternia sauna
nauplii were significantly larger in cones compared to larvae fed on the same
diet in cylindro-conical bottles (Tukey's HSD; p<O.O5; see Appendix A, Table 2). There
was no significant diet by container interaction in groups fed on MF only (Tukey's HSD;
p>O.O5; see Appendix A, Table 2).
There was a significant effect of diet on the survival of zebrafish larvae (ANOVA;
p=O.O374) with a mean survival of 73.7% for larvae in group MF+ART compared to
68.2% for larvae in group MF (see Appendix A, Table 3). In this experiment, no effect
of container type on the survival of zebrafish larvae was detected (Two-way ANOVA;
p=0.4938; Figure 8).
Throughout the experiment, water temperature was 28 °C except during the first 3
days. Due to unexpected sea water system maintenance and associated fluctuations of water
flow, sudden changes in temperature (28-22 °C) were observed. One-third of the total
mortality occurred during the first 3 days when temperature problems occurred. Oxygen
35
D MF
MF+ART
(5
>
(5
I..
.a
+
w
N
oE
a,
C
a,
-J
Type of culture vessel
Figure 7: Final ML of larval zebrafish at the end of 21 days. Larvae were fed on either of
two different diets in one of three different containers. Initial ML of larvae was 3.79 mm.
MF = Microfeast®, MFART = Microfeast®+ Arteinia sauna nauplii
36
MF
MF+ART
100
U)
6.
.0(l)
a)
N _
::FI1'
lmhoff
r
Cylinder
Cone
Rectangular
tank
Type of culture vessel
Figure 8 : Percent survival of larval zebrafish at the end of 21 days. Larvae were fed on
either of two different diets in one of three different containers. Initial ML of larvae was
3.79 mm. MF = Microfeast®, MF+ART = Microfeast®+Arternia sauna nauplii.
37
measurements were always above 7 ppm and pH was between 6.96-7.0 1. Total ammonia
levels were <0.2 ppm which was reported to be safe for non-salmonid freshwater fish.
Characteristics of Artificial Microparticles Used in Growth Experiments
with Zebrafish Larvae
Retention efficiency of CLPWC
Based on a standard curve the following equation described the relationship
between absorbance at 520 nm and concentration of Poly-R in distilled water (simple
regression; r2=99.98; see Appendix B, Table 1):
Absorbance = 0.000344873 + 11075.3 * concentration (mg/mi)
(12)
A second standard curve was obtained for the absorbances of different
concentrations of Poly-R in I M NaOH (simple regression; r2=99.98; see Appendix B,
Table 2). The equation of the fitted model was:
Absorbance = 0.00130092 + 14010.6 * concentration (mg/mI)
(13)
Table 3 summarizes the results obtained by using the equations above, for loss of
Poly-R from CLPWC after 2, 4 and 12 h of suspension of capsules in 10 ml distilled water
or I M NaOH. After 2 h of suspension in distilled water, the concentration of Poly-R in
distilled water was only 12.86% of that of PoIy-R in 1 M NaOH. After 12 h of suspension
in distilled water, it was found that the concentration of Poly-R dye lost from CLPWC was
only 18.88% of the concentration of Poly-R dye in I M NaOH. These results show that
even after a suspension period of 12 h, more than 80% of the encapsulated Poly-R could
be delivered to fish larvae by CLPWC.
Table 3 : Mean concentrations and percent Poly-R dye lost from CLPWC suspended in
either 10 ml dH2O or 10 ml 1 M NaOH after 2,4, and 12 h of suspension.
Sample time
(h)
Concentration
in
dH2O
(mg/I)
Concentration in
% Loss of PoIy-R
1 M NaOH (mg/I)
in dH2O
2
4.39*105
3.42*104
12.86
4
5.39*105
3.42*104
15.73
12
6.47*10.5
3.43*10.4
18.88
Retention efficiency of GAB
The estimated initial amount of Poly-Red in 10 mg of GAB was calculated to be
4.3 64 mg. The loss of Poly-R from of GAB was determined using equation (12) (page
37). Percent losses after 2, 4, and 12 h of suspension of GAB in distilled water are shown
in Table 4.
Table 4: Mean concentrations and percent loss of Poly-R dye after 2, 4, and 12 h
suspension of GAB in dH2O.
Sample Time
(h)
Concentration in
dH2O
(mg/I)
% Loss of PoIy-R in
dH2O
2
1.438*10.3
0.03%
4
2.641*10.3
0.06%
12
3.136*10.3
0.07%
The results indicated that after 2 hours of suspension, the concentration of Poly-R
dye in distilled water was 0.03% of the initial amount of Poly-R encapsulated in GAB.
After 4 hours and 12 hours of suspension in distilled water, it was determined that the
concentrations of Poly-R dye were 0.06 and 0.07% of the initial concentration of
39
encapsulated Poly-R in GAB, respectively. Since there was a loss of Poly-R dye during
the encapsulation process, it is likely that the initial amount of encapsulated Poly-R was
overestimated and the percent loss may, therefore, have been greater than estimated.
Acceptability of CLPWC and GAB
Acceptability of CLPWC and GAB by first feeding larvae
3.82 mm)
There were significant differences among the mean gut fullness values of starved
larvae and larvae fed on either MF, CLPWC or GAB (ANOVA; p<O.000l; see Appendix
C, Table 1). A Tukey's multiple range procedure for comparison of mean gut fullness
showed that larvae fed on MF had a significantly larger mean gut fullness (0.1 19±0.004
mm2) than that of larvae fed on CLPWC or GAB beads (0.098±0.002 mm2 and 0.102±
0.003 mm2, respectively; see Appendix C, Table 2). This difference in mean gut fullness
indicated a greater ingestion rate of MF compared to other diets. There were no significant
differences (Tukey; p>O.O5) in gut fullness of larvae fed on either CLPWC or GAB which
suggested that larvae ingested both CLPWC and GAB at similar rates (see Figure 9).
There were statistically significant differences in the mean feeding incidences of
zebrafish larvae fed on either CLPWC, GAB or MF (ANOVA; p< 0.0001; see Appendix
C, Table 3). The mean feeding incidences of larvae fed on CLPWC and GAB were found
to be 33% and 68%, respectively, compared to 94% for larvae fed on MF. Although the
mean gut fullness of larvae fed on either CLPWC or GAB were not significantly different
from each other (Tukey; p>O.O5), the mean feeding incidence of zebrafish larvae fed on
GAB was significantly greater than that of larvae fed on CLPWC (Tukey; p<O.O5; see
Appendix C, Table 4). Examination of the gut contents of larvae under the microscope
showed that the number of larvae that ingested GAB were twice that of larvae ingesting
CLPWC. On the other hand, MF was ingested at the greatest rate by zebrafish larvae (94%)
and this was significantly greater than the feeding incidence of larvae fed on GAB.
Gut fullness
S
Feeding Incidence
100
0.19
:80
E
o
0.161
,ci
60
_r40
EE1
0.13:
:
I
01
20
007
Starved
CLFWC
GAB
MF
Treatment
Figure 9: Mean gut fullness and feeding incidences of first feeding zebrafish larvae (5 day
old; ML 3.82 mm) fed on either MF, CLPWC or GAB. Larvae were fed on a mixture of
<20-75 .tm sized CLPWC or GAB in Imhoff cones (total volume I 200 ml). Stocking
density was 100 fish/cone with a flow rate of 4Oml/min.
41
The differences in gut fullness and feeding incidence of zebrafish fed on MF
showed that first feeding larvae, (MLS 3.8 mm) found MF to be more acceptable than
either CLPWC or GAB (Figure 9).
Acceptability of CLPWC and GAB by 14 day old larvae ( ML 4.24 mm)
There were significant differences among the gut fullness values of larvae fed on
either CLPWC, GAB, or MF and starved larvae (ANOVA; p<O.001; see Appendix D,
Table I). Comparisons of mean gut fullness values of the larvae showed that larvae fed
on GAB and MF had a mean gut fullness of 0.126±0.006 mm2 and 0.128±0.004 mm2,
respectively, and there was no significant difference between these two treatments (Tukey;
p>O.O5; see Appendix D, Table 2). However, larvae fed on CLPWC had a significantly
smaller mean gut fullness of 0.1 14±0.004 mm2, indicating a lower ingestion rate for these
capsules by larvae. The mean gut fullness of starved larvae was significantly smaller than
those of the other groups (gut fullness= 0.0961±0.003 mm2; Tukey; p<O.O5).
There were statistically significant differences in the mean feeding incidences of
zebrafish larvae fed on either CLPWC, GAB or MF (ANOVA; p<O.001; see Appendix D,
Table 3). The mean feeding incidences of larvae fed on GAB and MF were 94% and 96%,
respectively, and there was no statistically significant difference between these groups
(Tukey; p>O.O5). The mean feeding incidence of zebrafish larvae fed on CLPWC was
found to be 61% which was significantly lower (Tukey; p<O.O5) than that of larvae fed
on CLPWC or GAB (see Appendix D, Table 4).
When compared to the mean feeding incidences of 3.8 mm larvae, these results
showed that acceptability of GAB (both gut fullness and feeding incidence) increased with
larval age and was similar to that of larvae fed on MF (see Figure 10). Since the mean
particle size of GAB was larger than MF, this increase in acceptability could reflect an
increase in the particle size preference of larvae as they grew. However, acceptability of
42
Gut fullness
S
Feeding Incidence
0.15
i1III
0
C
(..1
E
C)
0
60
,CI)
.E
D)
0.12
C
40
U-
4-
ftn
Starved
CLPWC
GAB
20
fl
MF
Treatment
Figure 10: Mean gut fullness and feeding incidences of zebrafish larvae (12 day old; ML
4.24 mm) fed on either MF, CLPWC or GAB. Larvae were fed on a mixture of <20-106
tm sized CLPWC or GAB in Imhoff cones (total volume 1200 ml). Stocking density was
100 fish/cone with a flow rate of 4Oml/min.
43
CLPWC was still statistically lower than that of GAB and MF. This could be due to the
unpalatability or poor nutritional value of CLPWC which resulted in poorer acceptance by
zebrafish larvae.
Acceptability of CLPWC and GAB by 20 day old larvae (MLS 5.2 mm)
There were significant differences among the gut fullness values of larvae fed on
the different diets (ANOVA; p<O.0001; see Appendix E, Table 1). A Tukey multiple range
test showed that larvae fed on GAB had a significantly (p<O.O5) larger mean gut fullness
(0.310±0.004 mm2) than those fed on either MF or CLPWC (see Appendix E, Table 2).
Larvae fed on CLPWC had a mean gut fullness of 0.258±0.005 mm2 which was
significantly smaller than that of larvae fed on MF with a mean gut fullness of 0.284±0.003
mm2. Starved larvae had a mean gut fullness of 0.195±0.003 mm2 which was significantly
smaller (p<O.O5) than all other groups (see Figure 11).
Although the gut fullness values of larvae fed on different diets were significantly
different from each other, the feeding incidences of larvae was almost 100% for all diets
and there were no statistical differences among treatments (ANOVA and Tukey's HSD;
p>O.OS; see Appendix E, Tables 3 and 4). This high feeding incidence for larvae of 5.2
mm in length fed on CLPWC or GAB was greater than the feeding incidences of 3.8 and
4.2 mm larvae fed on the same diets. Although 5.2 mm larvae had a larger mouth gape
and were capable of ingesting brine shrimp nauplii, gut fullness of larvae fed on MF was
similar to that of larvae fed on GAB and was significantly greater than that of larvae fed on
CLPWC (Figure 11). The higher ingestion of MF by 5.2 mm larvae was unexpected as the
particle size of MF was much smaller (5-10 jtm) than GAB (<20-212 jtm). MF particles
were probably undetectable by the larvae compared to larger sized CLPWC and GAB. It
is possible that larger size zebrafish larvae, which are capable of ingesting brine shrimp
nauplii, can feed on 5-10 tm particles by filter feeding using their primordial gill slits.
Gut fullness
Feeding Incidence
---:
0.35
100
[80
0.3
C
ci
i60
C1)
0.25
..-
C,
40
+
4-
0.2
0.15
iiI;iLiL
Starved
CLPWC
GAB
MF
Treatment
Figure 11: Mean gut fullness and feeding incidences of zebrafish larvae (19 day old; ML
5.2 mm) fed on either MF, CLPWC or GAB. Larvae were fed on a mixture of <20-212
tm sized CLPWC or GAB in Imhoff cones (total volume 1200 ml). Stocking density was
100 fish/cone with a flow rate of 4Oml/min.
Digestibility of CLPWC and GAB
Both CLPWC and GAB were digested by first feeding larvae. Digestion of
CLPWC was obvious in the alimentary tract of larvae. A five-day old, first feeding
zebrafish larvae fed on CLPWC and sampled 3 h later is shown in Figure 12. The gut
was not coiled and was filled with CLPWC. Although capsules closer to the anterior
intestine were intact, the protein membranes of microcapsules were broken down in the
posterior intestine and released dye was evident.
Although GAB were irregular in shape the change in the color of the beads
indicated that GAB were digested by first feeding zebrafish larvae. Figure 13 shows the
difference in the color intensity between GAB in the foregut and those in the hindgut of a
zebrafish larva sampled 3 h after feeding. The color in the rectal area was pinkish-white
compared to intense pinkish color in the foregut area, indicating loss of encapsulated dye.
Furthermore, although it was not possible to distinguish the individual wall of GAB in
most cases, ingested material inside the digestive tract of larvae suggested that GAB were
physically broken down as shown in Figure 14.
In most cases, observation of the digestive tract of larvae is not possible for larvae
older than 10-15 days because the larval body lost its transparency due to pigmentation;
however, in the present study, observations of the digestive tract of zebrafish larvae up
to 20-25 days old were possible and showed that zebrafish larvae digested CLPWC and
GAB. Figure 15 shows fecal strands collected from 15 day-old larvae fed on CLPWC
after 3 h. The walls of CLPWC were not visible and the protein membranes of the capsules
were completely broken down. Free red dye was evident in the faeces of zebrafish larvae,
indicating that dye had been released from CLPWC. Digestion of GAB was more evident
in 14 day-old larvae as shown in Figure 16. The irregular shapes of ingested GAB were
not detectable in the guts of the larvae and the released dye was evident indicating break
down of GAB.
46
r.r
L-
Figure 12: Five day-old, first feeding zebrafish larva (lateral view) fed on CLPWC
containing a red dye (Poly-R 478). The protein membranes of CLPWC were broken down
in the posterior part of foregut and free red dye was evident in the gut of the larva.
=40 jim.
Magnification, 100X objective. Scale
47
Figure 13: Five day-old, first feeding zebrafish larvae (lateral view) fed on GAB containing
a red dye (Poly-R 478). Released red dye was visible indicating that Poly-R had leached
=40 pm.
from the digested beads in the gut. Magnification, 100X objective. Scale
48
Figure 14: Fecal strands collected from a 15 day-old zebrafish larva (lateral view) fed on
CLPWC containing a red dye (Poly-R 478). The protein membranes of capsules were
broken down and released red dye was evident. Magnification, 125X objective. Scale
=5O.tm.
49
Figure 15: Fourteen day-old zebrafish larva (lateral view) fed on GAB containing a red dye
(Poly-R 478). Red dye was visible indicating that Poly-R had leached from the digested
=80 .tm.
beads. Magnification, 50X objective. Scale
50
Size of CLPWC Selected by Zebrafish Larvae
Number of CLPWC/mg
The mean numbers of each size class of CLPWC per mg of capsules prepared for
this study are summarized below in Table 5.
Table 5: The mean numbers of CLPWC/mg of capsules prepared for this study (n=3).
Size class (jim)
Number of capsules/mg ± 1 SD
<20
235070±10850
21-45
67921±6056
46-75
7685±420
76-106
3079±125
107-212
402±26
213-425
88±5
Sinking Rate of CLPWC
The change in the mean concentrations of different size classes of CLPWC during
60 mm of suspension in Imhoff cones under flow rates of 40 or 80 ml/min are summarized
below in Tables 6 and 7.
There was a dramatic decrease in the concentration of CLPWC under a flow rate
of 40 mI/mm at the end of 1 h. This decrease in concentration was particularly significant
for capsules larger than 75 jim. When the flow rate was doubled, the greater up-welling
current increased capsule suspension, compared to that with the 40 ml/min flow rate. Since
suspension of capsules, particularly for size classes larger than 75 jim, was greater with
the 80 mI/mm flow rate, particle size preference experiments were carried out with a flow
51
Table 6: Mean number of CLPWC/ml measured at 15 mm intervals during a period of 60
mm under a flow rate of 40 mI/mm (n=3).
Number of capsules/mi
Size Class
(m
0
mm
15
mm
30
mm
± 1 SD
45
mm
60
mm
)
<20
10±0.5
8.4±0.1
5±0.6
3.4±0.3
2.4±0.3
21-45
10±0.9
8.2±0.6
5±0.7
3±0.6
2.2±0.3
46-75
10±0.4
6±0.2
2±0.3
0.8±0.1
0.4±0.6
76-106
10±0.7
2.8±0.1
0.6±0.1
0.4±0.1
0.4±0.1
107-212
10±0.6
1.4±0.4
0.2±0.1
0.2±0.6
0
Table 7: Mean number of CLPWC/mI measured at 15 mm intervals during a period of 60
mm under a flow rate of 80 mI/mm (n=3).
Number of capsules/mi
Size Class
(tm
0
mm
15
mm
30
mm
± 1 SD
45
mm
60
mm
)
<20
10±0.5
9±0.8
6±0.6
4.8±0.2
3±0.1
21-45
10±0.9
8.6±0.1
5.8±0.2
4.4±0.3
2.8±0.1
46-75
10±0.4
6.8±0.2
2.6±0.8
1.8±0.1
1.6±0.3
76-106
10±0.7
3.2±0.1
1.4±0.1
0.8±0.1
0.8±0.3
107-212
10±0.6
3±0.3
1.4±0.1
0.8±0.1
0.8±0.1
rate of 80 mllmin. Data in Table 7 were used to formulate an equation to describe the mean
sinking rate of CLPWC for each size class (equation (4); page 28). Predicted values were
then compared with experimental data (Figure 16 a through e).
52
The equation predicted that the concentrations of 76-106 and 107-212 im size
capsules after I h should be 4.1 and 4/mi showing a 59.3 and 60.0% decrease,
respectively, compared to initial concentrations. While the final concentration of 46-75
m size capsules should be 1.6/mi and the concentrations of 76-106 and 107-2 12 tim size
capsules should be both 0.8/mi, showing a decrease of 92% from the initial concentration.
Feeding experiments with zebrafish larvae
A given number of capsules of each size class were added to cones at 15 mm
intervals in order to compensate for predicted losses due to sinking and to maintain an
average concentration of 2 capsuies/ml for each size class during the particle size selection
experiment. The total number and size distribution of capsules ingested by zebrafish larvae
at the end of 60 mm were then determined for larvae of mean standard lengths of 3.8 or 5.2
mm (Table 8).
Table 8: Size distribution and total number of capsules in the foreguts of 3.8 and 5.2 mm
long zebrafish larvae fed on CLPWC for a period of 60 mm at a flow rate of 80 mi/mm in
Imhoff cones (n=lOO).
Size Distribution and Total Number
of Capsules in the Foregut of 100 Larvae
Fish Length
Fish Length
3.8 mm
5.2 mm
<20
21
17
21-45
394
707
46-75
196
1319
76-106
3
270
107-212
0
24
Capsule Size
(tim
)
53
a)
10
Predicted data
Experimental data
9
-.8
*
c6
0
E
I-
-.-
0
0
15
30
45
60
Time (mm)
b)
10
Predicted data
9
Experimental data
E
*
0
I-.
C
0
0
15
30
45
60
Time (mm)
Figure 16: Comparison of experimental data and predicted concentrations showing the rate
of loss of CLPWC during a period of 1 h, under a flow rate of 80 mi/mm. a) <20 im size,
b) 21-45 im size capsules.
54
c)
10
Predicted data
Experimental data
E
0
0
15
30
45
60
Time (mm)
d
10
Predicted data
9
Experimental data
E
70
I-
C
32
0
0
15
30
45
60
Time (mm)
Figure 16: Comparison of experimental data and predicted concentrations showing the rate
of loss of CLPWC during a period of I h, under a flow rate of 80 mi/mm. a) 46-75 tm
size, b) 76-106 tm size capsules.
55
e)
Predicted data
Experimental data
10
9
*
C6
0
C
C
0
0
0
15
30
45
60
Time (mm)
Figure 16: Comparison of experimental data and predicted concentrations showing the rate
of loss of CLPWC during a period of I h, under a flow rate of 80m1/min. e)107-212 j.tm
size capsules.
56
Determination of diet size preference of zebrafish larvae would be misleading
without knowing the concentration of capsules of each size class available to the larvae
throughout the experiment. Figure 17 shows a comparison of the predicted mean
concentrations of differently sized capsules in the water column in two different
experiments with 3.8 and 5.2 mm larvae after allowing for both sinking losses and losses
due to ingestion of capsules by larvae. In both experiments capsules <20 j.tm were present
at the greatest concentration indicating poor ingestion by larvae. The mean concentrations
of 107-2 12 pm capsules were similar in both experiments, suggesting that zebrafish
larvae of 3.8 and 5.2 mm did not ingest this size range of capsule. On the other hand, the
predicted mean concentration of 76-106 jtm capsules in the water column in the experiment
with 5.2 mm larvae was slightly lower than with 3.8 mm larvae. Although the mean
concentration of 46-75 tm sized capsules in the water column was the least in both
experiments with 3.8 and 5.2 mm larvae, analysis of gut contents showed that 3.8 mm
larvae preferred capsules in the 2 1-45 jim size range.
In this study, diet size selection of zebrafish larvae was determined by considering
the concentrations of different size capsules both in the foreguts of larvae and in the water
column (equation (6); (page 30). Diet size selections are summarized for larvae of 3.8 and
5.2 mm length in Tables 9 and 10, respectively.
Analysis of the percent relative particle size selection for 3.8 mm zebrafish larvae
showed that 21-45 j.tm capsules were ingested by larvae with a relative selection value of
63.5%, compared to 3.1% and 32.9% for <20 and 46-75 jim capsules, respectively.
Larvae were not capable of ingesting 107-2 12 jim capsules during first feeding.
Zebrafish larvae 5.2 mm in length preferred larger sized particles compared to 3.8
mm larvae. Relative diet size selection for 5.2 mm zebrafish larvae indicated low preference
(0.7%) for 21-45 jim capsules. On the other hand, 46-75 jim capsules were preferred (61%
relative selection) compared to 2 1-45 jim capsules (27.2%). Although 5.2 mm larvae
ingested 107-2 12 jim capsules, these large capsules was least preferred (0.9%).
57
D Mean Length=3.8 mm D Mean Length=5.2 mm
1.75
a.
U)
1.5
0
.-
OC
C
U
C
0
0
-
fl_QQQQQI --
1
<20
I.r.a00000. -
--
-
_
21-45 46-75 76-106 107-212
Size class
Figure 17: Comparison of the predicted concentrations of different size classes of CLPWC
suspended in the water column, during 60 mm with larvae of either ML, of 3.8 or 5.2 mm
after accounting for the effects of both sinking and capsule ingestion by larvae.
Table 9: Diet size selection of zebrafish larvae (MLS 3.8 mm) fed on CLPWC during 60
mm with a flow rate of 80 ml/min.
Size class Total
(m
)
#
of capsules in
foregut of all larvae
(100
larvae/1200
ml:
Mean particle
concentration
(#/1200
Selection
%
coefficient Relative
selection
ml)
<20
21
1851
0.011
3.1
21-45
394
1745
0.226
63.5
46-75
196
1673
0.117
32.9
76-106
3
1720
0.002
0.5
107-212
0
1729
0.000
0
Table 10: Diet size selection of zebrafish larvae (MLS 5.2 mm) fed on CLPWC during 60
mm with a flow rate of 80 ml/min.
Size class Total# of capsules in
(tm
)
foregut of all larvae
(100
larvae/1200m1)
Mean particle
Selection
%
concentration
coefficient Relative
ml)
selection
(#11200
<20
18
1852
0.010
0.7
21-45
707
1631
0.433
27.2
46-75
1319
1360
0.970
61.0
76-106
270
1666
0.162
10.2
107-212
24
1724
0.014
0.9
59
Growth and Survival of Zebrafish Larvae Fed On CLPWC or GAB as a
Substitute for Brine Shrimp, Artemia sauna
Growth and survival of zebrafish larvae fed on CLPWC as a substitute for Artemia sauna
The initial ML, of the experimental larvae was 3.79±0.03 mm. The ML, and mean
survival of the larvae at the end of the 12 days of pre-conditioning were 5.14±0.06mm
and 96±1 .95%, respectively, and there was no significant difference among larvae in any
of the cones (ANOVA; p=O.933 and 0.7365; see Appendix F, Tables 1 and 2). At the end
of 21 days, after feeding the fish on diets substituted with CLPWC for 8 days, there were
significant differences among treatment groups (ANOVA; p<O.000l; see Appendix F,
Table 3).
Growth of zebrafish larvae in different treatments were compared by a Tukey's
multiple comparison test (p<O.O5) to determine which means were different from others.
There was no significant difference between the final lengths of larvae fed on 100%
CLPWC and 100% MF and these two groups showed the smallest growth with final
ML, of 5.27 mm and 5.29 mm, respectively. Larvae fed on 80% CLPWC+20% ART had
a ML, of 6.15 mm which was significantly greater than that of larvae fed on 100% CLPWC
and MF but significantly smaller than that of larvae fed on 60% CLPWC+40% ART (ML,
6.47 mm). There were no statistical differences between the lengths of larvae fed on 40%
CLPWC+60% ART, 20% CLPWC80% ART or 100% ART (see Appendix F, Table 4).
There were significant survival differences among treatment groups (ANOVA;
p<O.000l; see Appendix F, Table 5). Further analysis with a multiple range test using
Tukey's multiple range tests revealed that zebrafish larvae fed on 100% CLPWC had the
lowest survival rate of 43%, which was significantly lower than that of larvae fed on 80%
CLPWC+20% ART or 60% CLPWC+40% ART with mean survival rates of 64% and
68%, respectively (p<O.O5). There were no statistically significant differences among
survival of groups fed on 40% CLPWC+60% ART, 100% MF, 20% CLPWC+80% ART
7.14
ri
0
j
d
6.64
C
1614
n
5.64
a
5 14
L
a
ri JL
ME
-
0
20
40
60
80
100%
Artemia
100
80
60
40
20
0 %
CLPWC
Dietary
treatment
Figure 18: Comparison of mean lengths of zebrafish larvae fed on Artemia nauplii
substituted with different levels of CLPWC at the end of 8 days. The larvae were 17 days
old initially and the initial ML was 5.14 mm. Letters denote significant differences
(Tukey's Honest Significant Difference (HSD) multiple range test; p<O.O5).
61
85
a
-
a
75
65i
ii
Fiti
I
45
b
35
n
MF
0
20
40
60
80
100
80
60
40
20
Dietary
100% Artemia
0%
CLPWC
treatment
Figure 19: Comparison of mean survival of zebrafish larvae fed on Artemia nauplii
substituted with different levels of CLPWC at the end of 8 days. The larvae were 17 day
old initially and the initial ML was 5.14 mm. Letters denote significant differences
(Tukey's Honest Significant Difference (HSD) multiple range test; p<O.O5).
62
or 100% ART at rates of 69.6%, 70.3%, 72.6%, and 74.3%, respectively (see Appendix
F, Table 6).
Throughout the experiment, water quality parameters were as follows: water
temperature was 28±1 °C. Oxygen measurements were between 6.98-7.85 ppm and
pH was between 6.95-7.09. Total ammonia levels were <0.2 ppm. Bacterial layers and
uneaten food on the sides of the containers were removed by cleaning every 3 days with
a glass pipette.
Growth and survival of zebrafish larvae fed on GAB as a substitute for Artemia sauna
The initial ML of the experimental larvae was 3.81±0.03 mm. The ML and
survival of larvae at the end of the 12 days of pre-conditioning were 5.22±0.05 mm and
96±1 .45%, respectively and there were no significant differences in growth (p=O.539) and
survival (p=O.644) among larvae in any of the cones (ANOVA; see Appendix G, Tables I
arid 2). At the end of 8 days, after feeding the larvae on Arternia nauplii substituted with
different levels of GAB, there were statistically significant differences in the final ML of
larvae in the different treatments (ANOVA; p<O.000I; see Appendix G, Table 3). In order
to determine which means were significantly different from others, Tukey's multiple range
test was applied (p<O.OS; see Appendix G, Table 4).
Larvae fed on 100% GAB had a final ML of 5.87 mm which was statistically less
than that of larvae fed on 100% MF and 80% GAB+20% ART with final mean lengths of
6. 195 and 6.267 mm, respectively. Moreover, there was no statistically significant
difference between the final ML of larvae fed on 100% MF and 80% GAB+20% ART
(Tukey; p>O.OS). The final ML, of larvae fed on 60% GAB+40% ART (6.633 mm) was
significantly greater (p<O.O5) than that of larvae fed on 100% MF and 80% GAB+20%
ART but was significantly less than that of larvae fed on 40% GAB+60% ART (7.184
mm). There was a significant difference in the final ML, of larvae fed on 100% ART and
63
40% GAB+60% ART; however, there were no significant difference in the final lengths of
larvae fed on 100% ART and 20% GAB80% ART.
There was a statistically significant difference in the mean survival of larvae from
different treatments (ANOVA; p=O.0059; see Appendix G, Table 5). A Tukey's multiple
range procedure was used to discriminate among the mean survivals of the larvae (see
Appendix G, Table 6). The mean survival of larvae fed on 100% GAB was 92.3% but it
was significantly less (p<O.O5) than that of larvae fed on 80% GAB20% ART or 40%
GAB+60% ART, with mean survivals of 98% and 98.6%, respectively. Although there
were slight differences in survival among the other groups, these differences were not
statistically significant, with the lowest larval survival of 94.3% occurring with a diet of
60% GAB+40% ART and the highest larval survival of 98.65% occurring with a diet of
40% GAB+60% ART (p>O.O5).
Throughout the experiment water quality parameters were similar to the previous
experiment. Water temperature was 28±1 °C. Oxygen measurements were between 7.15-
7.95 ppm and pH was between 6.92-7.19. Total ammonia levels were <0.2 ppm. Bacterial
layers and uneaten food on the sides of the containers were removed by cleaning every 3
days with a glass pipette.
e
7.72
d
ci
7.22]
- - -
I
- - - -
:::
n ñ
MF
0
20
40
60
80
100
80
60
40
20
Dietary
100% Artemia
0%
GAB
treatment
Figure 20: Comparison of mean lengths of zebrafish larvae fed on Arteinia nauplii
substituted with different levels of GAB at the end of 8 days. The larvae were 17 days old
initially and the initial ML was 5.22 mm. Letters denote significant differences (Tukey's
Honest Significant Difference (HSD) multiple range tests; p<O.05).
a
100
a
98
a
a 96
a
ui
+
92
90
88
c
86
84
:
I
82
80
L
MF
0
20
40
100
80
60
Dietary
U
80
20
ft
100% Artemia
0% GAB
treatment
Figure 21: Comparison of mean survival of zebrafish larvae fed on Arternia nauplii
substituted with different levels of GAB at the end of 8 days. The larvae were 17 days old
initially and the initial ML was 5.22 mm. Letters denote significant differences (Tukey's
Honest Significant Difference (HSD) multiple range tests; p<O.OS).
DISCUSSION
Effects of Diet and Container Type on the Growth and Survival of
Zebrafish Larvae
Poor growth and survival of zebrafish larvae fed on artificial diets have been
explained in many ways as discussed in the introduction. The morphological, functional
and physiological properties of the larval alimentary tract have been considered, as has the
nature of artificial diets. In the present study, larvae >12 days old that were fed on Arternia
sauna grew faster than those fed on MF only. The poor growth and survival of larvae fed
on MF could be due to the small particle size of MF that prevents visual capture of food
particles. It has been reported that dissolved substances and particles smaller than 20 1m
were not nutritionally available to cyprinid larvae (Kamler, 1992). In addition to diet
particle size, leaching of nutrients could have contributed to the poor growth and survival
of zebrafish larvae fed on MF. Nevertheless, in the present study, it was shown that MF
supported growth and survival of zebrafish larvae during the first 12-14 days of exogenous
feeding, until the larvae were capable of ingesting larger size live food organisms. MF was
readily available and feeding the larvae with MF was less labor intensive compared to the
use of other diets such as paramecia and other infusoria as reported by Westerfield (1993).
Although there was no statistically significant difference in growth and survival
rates of zebrafish larvae fed on MF alone among the three types of containers, the probable
explanation for higher growth of larvae fed on MFART in conical shaped containers
was a better flow pattern that led to prolonged suspension of Artemia sauna in the water
column. it was observed that the up-welling flow pattern in cones allowed larvae to feed
on prey organisms with less swimming activity and prey items stayed in suspension
for longer periods of time. Upweiling currents also ensured that the movement of food
particles attracted the larvaes attention. On the other hand, more static conditions in
rectangular containers enabled larvae to search for food easier but the prey organisms
67
were not available in the water column due to shorter depth of these containers and rapid
particle sinking rates. Other methods have been described specifically for larval culture
of zebrafish including rearing larvae in beakers and acrylic tubes (Westerfield,
1993).
However, intensive rearing of zebrafish larvae in Imhoff cones with a flow-through system
is easy and based on the results of this study, results in improved larval performance.
Oxygen, pH and ammonia levels in the containers throughout the experiment were
optimal for the optiomal growth and survival of zebrafish larvae. Larvae suffered from
temperature fluctuations experienced during the first three days of the experiment and were
observed to have air bubbles inside their alimentary tracts. This problem was irreversible
and all the larvae with air bubbles died within couple of days, accounting for one-third of
the total mortality. After the third day, the temperature stabilized and no further fluctuations
were observed throughout the rest of the experiment. Although zebrafish larvae seemed to
successfully adapt to different culture temperatures, maintaining a stabile temperature
is crucial for good survival and growth rates of larvae. Although there were no reports
available for the temperature-specific growth rates of zebrafish larvae, our experience
has shown that 28°C produced rapid growth and all experiments were carried out at this
temperature.
Characteristics of Artificial Microparticles Used in Growth Experiments
with Zebrafish Larvae
Retention efficiency of CLPWC and GAB
Leakage rates for CLPWC tested in this study were low (<20% ) after
suspension in distilled water and were similar to leakage rates for CLPWC
containing protein (Langdon,
1989).
12
h of
(5-20%)
These leakage rates suggest that CLPWC used in
this study ensure availability of dietary nutrients to zebrafish larvae during prolonged
suspension of food particles in the water column. Although it was not possible to determine
the true initial concentration of encapsulated core material in experiments with GAB,
percent loss values after 12 h of suspension in distilled water suggested that GAB can
deliver high molecular weight, water soluble nutrients to freshwater fish larvae without
considerable losses.
Acceptability of CLPWC and GAB
Zebrafish larvae were able to capture and ingest CLPWC and GAB from first
feeding onwards. Both gut fullness and feeding incidence of fish larvae were superior with
GAB compared to those for larvae fed on CLPWC for all larval sizes examined, except that
the feeding incidence for 5.2 mm zebrafish larvae fed on CLPWC and GAB were the same
and close to 100%. Acceptance of microencapsulated feeds from the onset of exogenous
feeding has been previously reported for marine fish larvae; for example, higher acceptance
rates of CLPWC by seabass, Lates calcarjfer, and alginate microbound diets by European
seabass, Dicentrarchus labrax, were reported by Walford etal., (1991) and Person Le
Ruyet et al., (1993), respectively.
Although MF had a mean particle size of 5-10 .tm, acceptance of MF remained
higher for 4.2 and 5.2 mm larvae compared to CLPWC and GAB. Although it has been
suggested that particles smaller than 20 tm are not available for ingestion by cyprinid
larvae (Kamler, 1992) the high acceptance of MF by zebrafish larvae could have been due
to different capture mechanisms from those based on visual predation, i.e. by filtration of
particles by the gill slits. This recent theory has been studied in larvae of cod, Gadus
,norhua, and Atlantic halibut, Hippo glossus hippo glossus using algae as a first food (Van
Der Meeren, 1991; Reitan et al., 1994). These authors concluded that filter feeding on algae
which are known to contain high concentrations of free amino acids, may play an important
nutritional role by providing dissolved essential nutrients for early feeding stage larvae.
Uptake of dietary particles by filter feeding, involving primordial gill slits, may be an
adaptive mechanism for zebrafish larvae when encountered small sized diet particles.
Digestibility of CLPWC and GAB
Observations of the alimentary tract of zebrafish larvae showed that CLPWC were
digested by first feeding zebrafish larvae. When the larvae were sampled 1/2-1 h after
feeding on CLPWC, larval guts contained more capsules compared to those sampled 3-4
h later but there was less capsule digestion, as indicated by partially broken down protein
membranes that was still visible through transparent bodies of larvae. On the other hand,
larvae sampled 3-4 h after feeding were observed to have fewer capsules in their guts but
the degree of digestion improved, as indicated by the amount of dye released by digested
capsules and the absence of visible microcapsule protein membranes. In larvae that were
capable of ingesting Artemia sauna nauplii, the gut passage time of CLPWC was shortened
and the larvae digested the capsules in a shorter period of time compared to first feeding
larvae. In a similar study with CLPWC, Walford et al., (1991) showed that seabass, Lates
ccilcarzfer, larvae up to the 8th day after hatching, could not digest this capsule type when
they were fed alone, suggesting insufficient digestive enzyme activity of early larvae. The
same authors also reported that CLPWC were broken down and adsorbed in the rectal area
of the guts of larvae when these capsules were fed to larvae in combination with rotifers. In
this study, however, first feeding (5 day-old) zebrafish larvae of were able to digest
CLPWC when they were fed only CLPWC.
Observations of the alimentary tract of first feeding zebrafish larvae showed that
GAB were digested. Individual GAB were difficult to observe inside the gut because they
were irregular in shape. Nevertheless, once ingested by first feeding larvae, GAB changed
shape and the walls were not visible in the hindgut area suggesting that GAB were broken
down, releasing the red dye inside the gut. It was also observed that when CLPWC and
GAB reached the narrower hindgut area, peristaltic movements in this area helped the break
down of particles, particularly larger sized ones. With respect to this last point, it seems
important to feed larvae with particles that are large enough to be physically broken down
in the hindgut area but small enough not to cause any blockage of this area.
70
Size of CLPWC Selected by Zebrafish Larvae
Sinking rates of CLPWC
In this study, an upwelling flow rate of 80 mi/mm in the cones provided greater
suspension microparticles in the water column compared to a flow rate of 40 mi/mm. Flow
rates higher than 80 mI/mm were not possible in this study because 80 mI/mm was the
maximum flow rate possible for the larval rearing system. Although some of the smaller
sized capsules were observed to adhere to the sides of the container, in general, CLPWC
larger than 75 jtm sunk to the bottom in a relatively short period of time, and less than 20%
of capsules larger than 75 tm were available for the larvae after 30 mm. The high sinking
rates of capsules larger than 75 JIm can result in decreased ingestion of capsules by larvae
and could have impaired growth. Suspension of CLPWC in the water column may be
prolonged by adding lipids to the particles.
Feeding experiments with zebrafish larvae
Many species of larvae ingest their prey whole, so that the upper limit of acceptable
prey size is determined by mouth width. In general, larval mouth width increases in
proportion to both larval length and the maximum size range of prey. Heath (1992)
suggested that the lower limit of acceptable prey size appears to be partly determined by the
metabolic requirements of the larvae whilst the upper limit is determined by mouth width.
Heath (1992) reported that larvae reared on only one size class of prey eventually grow to a
size where the metabolic cost of search and capture exceeds the calorific value of the small
prey item. In this study, feeding zebrafish larvae on equal concentrations of different size
particles indicated which particle size was preferred by larvae.
From the start of exogenous feeding, zebrafish larvae accepted a wide range of
particle sizes, but larvae preferred particles much smaller than the maximum size they could
ingest. Furthermore, it was determined that the preference for 76-106 tm capsules by 5.2
71
mm larvae was 20 times greater than that for 3.8 mm larvae, suggesting that diet size
preference of larvae is directly related to larval length, as reported by Detwyler and Houde
(1970), Stepien (1976), Polo et al. (1992) and Fernandez-Diaz et al. (1994) for other larval
fish species.
Although 76-106 tm capsules represented only a small proportion of the total
ingested particles, in terms of weight, they would account for a considerable proportion of
the total ingested biomass in the gut. In this study, while the upper size limit of capsules
that could be ingested by first feeding zebrafish larvae was determined to be around 76 J.tm,
this limit increased to more than 107 tm for larvae 5.2 mm in length. These results suggest
that zebrafish larvae can ingest particles much larger than the preferred size when they
encounter a wide size range of particles.
Fernandez-Diaz et al. (1994) suggested that the chance of any organism to be a
p:)tential prey depends on the physical capacity of predatory larvae to catch and eat it. The
same authors working with gilthead seabream, Sparus aurata, suggested that first feeding
larvae could not ingest Artemia sauna nauplii but started positively selecting these
organisms once the physical constraint had been overcome although the ratio of prey
widthlmouth width was not optimal. This indicates the importance of physical
characteristics of diet particles as mentioned in the previous section. Although zebrafish
larvae of 5.2 mm ingest a wide size range of particles, the maximum size range of the
capsules ingested was smaller than the size of Artenzia nauplii which they are normally fed
on by culturists. This suggests that deformability of diet particles may help larvae ingest
larger size particles.
72
Growth And Survival of Zebrafish Larvae Fed On CLPWC or GAB as a
Substitute for Brine Shrimp, Artemia sauna
Growth and survival of zebrafish larvae fed on CLPWC as a substitute for brine shrimp.
Artemia sauna
The results showed that up to 40% substitution of brine shrimp by CLPWC could
be accomplished without significantly retarding growth and survival of zebrafish larvae
after a feeding period of 8 days. Further substitutions of brine shrimp by CLPWC resulted
in decreases in either growth and survival rates of zebrafish larvae. Feeding larvae on
100% MF or 100% CLPWC resulted in poor growth rates and in the case of 100%
CLPWC, poor survival rates compared to other groups. Two factors could have
contributed to the poor growth of larvae fed on 100% MF:
1) the smaller particle size of MF allowed larvae to grow to only a certain size,
above which the metabolic cost of search and capture exceeded the calorific value of
ingested MF.
2) the nutritional value of MF was not sufficient for optimal growth of zebrafish
larvae due to lack of essential nutrients or due to rapid leaching of these nutrients after
suspension in water.
Inferior growth and survival of larvae fed on 100% CLPWC was probably due to
poor breakdown and incomplete digestion of CLPWC by zebrafish larvae without a fully
developed digestive system. Although all the larvae were fed on Arternia sauna nauplii for 2
days before feeding on CLPWC, inferior growth and survival rates of larvae fed on brine
shrimp substitutions greater than 40% suggested that CLPWC were only partially useful
as an Arteniia substitute. The requirement for >60% Artemia in mixed CLPWC/Artemia
diets may be the result of the need for exogenous supplements of digestive enzymes
from ingested prey. These enzymes may provide a substantial contribution to the total
enzymatic activity of the larval alimentary tract and could help break down the protein walls
of capsules. Furthermore, it was reported that high consumption of artificial diets and
accelerated gut evacuation rate leads to dilution of digestive enzymes, reduced reabsorption
73
in the hindgut, losses of proteins, decreased growth, and increased mortality in roach,
Rutilus rutilus, larvae (Kamler, 1992).
Growth and survival of zebrafish larvae fed on GAB as a substitute for brine shrimp.
Artemia sauna
Results showed that 20% substitution of Artemia nauplii by GAB can be achieved
without retarding growth and survival of zebrafish larvae. Greater substitutions of Arternia
nauplii by GAB resulted in reduced growth but not survival except when larvae were fed
on 100% GAB. The growth and survival of zebrafish larvae fed on 100% GAB was
inferior compared to that of larvae fed on 100% MF, although the larvae fed on the former
were exposed to larger size diet particles. This result could have been due to lack of
exogenous enzymes capable of breaking down GAB in the alimentary tract of zebrafish
larvae. In growth studies with European sea bass larvae fed on alginate microbound diets,
Person Le Ruyet et al. (1993) showed that in comparison with a live diet, 30% growth
reduction occurred and production was unreliable both in terms of growth and juvenile
quality. In the present study, although the larvae were observed to consume all Arteinia
nauplii shortly after adding them to containers every hour, a change in stomach color from
orange to yellowish-white by the end of every hour showed that larvae actively fed on
GAB once they consumed all Artemia nauplii.
74
CONCLUSION
Experiments described in this study demonstrated that conical shaped Imhoff cones
and an upwelling type of flow increased suspension time of food particles in the water
co'umn which in turn ensured their greater availability to fish larvae. In addition, frequent
feeding helped maintain food particle concentrations.
Since the acceptance of MF was high and MF promoted growth and survival of
zebrafish larvae over the first 10-12 days, MF could be used as a control diet in the
evaluation of microparticulate diets. However, for larvae that are capable of ingesting live
food organisms, a comprehensive evaluation of microparticulate diets can only be
accomplished by comparing growth and survival of larvae fed on microparticulate diets
with those fed on live food, such as Artemia
sauna.
Experiments with microparticulate diets showed that acceptance of GAB was higher
compared to that of CLPWC. The sinking rates of CLPWC above 75 jim were high and
capsules larger than 75 were not available to larvae after 15-30 minutes. Further studies
are needed to increase suspension and acceptability of CLPWC. Both suspension and
acceptability of diets can be increased by encapsulating different ingredients such as
lipids and other attractants. Stabilities of CLPWC and GAB were promising and suggested
that high molecular weight, water soluble nutrients could be delivered to freshwater fish
larvae without considerable losses. Experiments with microparticulate diets showed that
both CLPWC and GAB were digested by first feeding zebrafish larvae and digestion
was even more evident in older larvae. The degree of digestion also depended on the
concentration of capsules in the water column. High consumption rates decreased
gut passage time of capsules which in turn reduced digestion efficiencies. Therefore,
an optimal concentration of capsules should be maintained in the water column in
order to optimize digestion of capsules and to maintain growth. These experiments
demonstrated that although some of the problems related to rearing techniques can partially
75
be overcome by modifications in culture vessels, water flow and feeding methods, intrinsic
characteristics of microparticulate diets should be optimized to ensure satisfactory growth
and survival of fish larvae during the vulnerable early period of development.
Experiments on size selection of CLPWC by zebrafish larvae demonstrate that diet
size preference is a function of larval length. Although zebrafish larvae ingested a wide size
range of CLPWC, the maximum size of ingested particles was less than the size of Artemia
nauplii which 5.2 mm larvae are capable of ingesting. This result suggested that physical
characteristics of particles, such as deformability, could allow fish larvae to ingest large
particles.
In this study, it was demonstrated that up to 40 and 20% substitutions of Artemia
sauna nauplii by CLPWC and GAB, respectively, were possible without any reduction in
growth and survival of zebrafish larvae. A possible explanation for the inferior growth
was the incomplete nutrient composition of the microparticulate diets. It seems likely that
growth and survival of zebrafish larvae fed on CLPWC and GAB can be improved and
higher substitutions of Artemia nauplii can be obtained by improving the nutritional
composition and physical characteristics of CLPWC and GAB. However, a comprehensive
evaluation of microparticulate diets will only be possible after the specific nutritional needs
of the larvae are determined.
BIBLIOGRAPHY
Baragi, V. and Lovell, R. T., 1986. Digestive enzyme activities in striped bass from first
feeding through larval development. Transactions of the American Fisheries
Society, 115: 478-484.
Bengtson, D. A., 1993. A comprehensive program for the evaluation of artificial diets.
Journal of the World Aquaculture Society, 24: 285-293.
Buchal, M. A., 1994. Microencapsulation of water-soluble substances for delivery to
marine bivalves. Master Thesis, Oregon State University, pp. 48-50.
Cahu, C.L. and Zambonino Infante, J.L., 1994. Early weaning of seabass (Dicentrarchus
labrax) larvae with a compound diet: effect on digestive enzymes. Comparative
Biochemistry and Physiology, 109A: 2 13-222.
Cahu, C. L. and Zambonino Infante, J. L., 1997. Is the digestive capacity of marine fish
larvae sufficient for compound diet feeding? Aquaculture International, 5: 15 1-160.
Chang, F. M. S., Machintosh, F. C. and Mason, S. C., 1966. Semi-permeable aquaeous
microcapsules. I. preparation and properties. Canadian Journal of Physiology and
Pharmacology, 44: 115-128.
Coves, D., Dewavrin, G., Breuil, G. and Devauchelle, N., 1991. Culture of seabass
(Dicentrarchus labrax), pages 289-303 in McVey, editor, CRC Handbook of
Mariculture, Boca Raton, Florida, USA.
Creswell, R., L., 1993. Aquaculture desk reference. Von Nostrand Reinhold Publishers,
pp. 68-75.
Dabrowski, K. R., 1979. Role of proteolytic enzymes in fish digestion. Pages 107-126 in
E. Styczynska-Jurewicz, E. Backiel, T. Jaspers, and E. Persoone, editors.
Cultivation of fish fry and its live food. European Mariculture Society, Special
Publication 4, Bredene, Belgium.
Detwyler, R. and Houde, E. D., 1970. Food selection by laboratory-reared larvae of the
scaled sardine Harengula pensacolea (Pisces, Clupoidae) and the bay anchovy
Anchoa initchilli (Pisces, Engraulidae). Marine Biology, 7: 2 14-222.
Fernández-Diaz, C., Pascual, E. and Ytifera, M., 1994. Feeding behaviour and prey size
selection of gilthead seabream, Sparus aurata , larvae fed on inert and live food.
Marine Biology, 118: 323-328.
77
Govoni, J. J., Boehiert, G. W. and Watanabe, Y., 1986. The physiology of digestion in
fish larvae. Environmental Biology of Fishes, 16: 59-77.
Heath, M. R., 1992. Field investigations of the early life stages of marine fish. Advances
in Marine Biology, 28: 1-153.
Holt, G. J., 1993. Feeding larval red drum on microparticulate diets in a closed
recirculating water system. Journal of the World Aquaculture Society, 24: 225-230.
Jones, D. A., Munford, J. G. and Gabbott, P. A., 1974. Microcapsules as artificial
particles for aquatic filter feeders. Nature (Landon), 247: 233-235.
Jones, D. A., Kamarudin, S. M. and Le Vay, L., 1993. The potential for replacement of
live feeds in larval culture. Journal of the World Aquaculture Society, 24: 199-211.
Kamler, E.,1992. Early life history of fish: an energetics approach. Chapman & Hall
Publishers, London, pp. 211-215.
Kanazawa, A., Koshio, S. and Teshima, S., 1989. Growth and survival of larval red
seabream Pagrus major and Japanese flounder Paralichthys olivaceus fed
microbound diets. Journal of the World Aquaculture Society, 20: 31-37.
Kolkovski, S., Tandler, A., Kissil, G. Wm. and Gertler, A., 1993. The effect of dietary
exogenous digestive enzymes on ingestion, assimilation, growth and survival of
gilthead seabream (Sparus aurata, Sparidae, Linnaeus) larvae. Fish Physiology and
Biochemistry, 12: 203-209.
Kreeger, D. A. and Langdon, C. J., 1993. Effect of dietary protein content on growth of
juvenile mussels, Mytilus trossulus (Gould 1850). Biological Bulletin, 185: 123139.
Langdon, C. J., 1983. New techniques and their application to studies of bivalve nutrition.
In: G. D. Prider, C. J. Langdon and D. E. Conklin, editors, Proceeding of The
Second International Conference on Aquaculture Nutrition. Louisiana State
University, Baton Rouge, pp. 305-320.
Langdon, C. J. and Bolton, E. T., 1984. A microparticulate diet for a suspension-feeding
bivalve mollusc, Crassostrea virgin ica (Gmelin), Journal of Experimental Marine
Biology and Ecology, 82: 339-258.
Langdon, C. J. and Siegfried, C.A., 1984. Progress in the development of artificial diets
for bivalve filter feeders. Aquaculture, 39: 135-153.
Langdon, C. J., Levine, D. M. and Jones D. A., 1985. Microcapsulated feeds for marine
suspension feeders. Journal of Microencapsulation, 2: 1-1 1.
Langdon, C. J., and Waldock, M. J. (1981). The effect of algal and artificial diets on the
growth and fatty acid composition of Crassostrea gigas spat. Journal of the Marine
Biological Association of the United Kingdom, 66: 431-448.
Langdon, C. J., 1989. Preparation and evaluation of protein microcapsules for marine
suspension-feeders, the Pasific oyster, Crassostrea gigas. Marine Biology, 102:
217-224.
Lauf, M. and Hoffer, R., 1984. Proteolytic enzymes in fish development and the
importance of dietary enzymes. Aquaculture, 37: 335-346.
Lemm, C. A. and Lemarie, D. P., 1991. Survival and growth of larval striped bass
(Morone saxatilis ) fed Arternia enriched with highly unsaturated fatty acids
(HUFA). Aquaculture, 99: 117-126.
I..evine, D. M., Sulkin, S.D. ans Van Heukelem, L., 1983. The design and development
of microencapsulated diets for the study of nutritional requirements of brachyuran
crab larvae. In the Culture of Marine Invertebrates: Selected Readings, edited by
C.J. Berg Jr. Hutchinson Ross, pp. 193-204.
López-Alvarado, J., Langdon, C. J., Teshima, S., Kanazawa, A., 1994. Effects of
coating and encapsulation of crystalline amino acids on leaching in larval feeds.
Aquaculture, 122: 335-346.
Molven, A., Njolstad, P. R. and Fjose, A., 1993. Pattern formation in fish embryos:
Characterization of homeobox genes from the zebrafish (Brachydanio re rio ). In B.
T. Waither and H.J. Fyhn, editors, Physiological and Biochemical Aspects of Fish
Development. University of Bergen, Norway, pp. 30-39.
O'Connell, C. P., 1981. Development of organ systems in the northern anchovy,
Engraulis mordax, and other teleosts. American Zoologist, 21: 429-446.
Ozkizilcik, S. and Chu, F. E., In press. Preparation and characterization of a complex
microencapsulated diet for striped bass, Morone saxatilis larvae.
Pedersen, B. H., Nilssen, E. M. and Hjelmeland, K., 1987. Variations in the content of
trypsin and trypsinogen in larval herring (Clupea harengus ) digesting copepod
nauplii. Marine Biology, 94: 171-181.
79
Person Le Ruyet, J., Alexandre, J. C., Thébaud, L. and Mugnier, C., 1993. Marine fish
larvae feeding: Formulated diets or live prey? Journal of the World Aquaculture
Society, 24: 211-224.
Polo, A., Yüfera, M. and Pascual, E., 1992. Feeding and growth of gilthead seabream,
(Sparus aurata L.) larvae in relation to the size of the rotifer stain used as food.
Aquaculture, Amsterdam, 103: 45-54.
Reitan, K. I., Bolla, S. and Olsen, Y., 1994. A study of the mechanism of algal uptake in
yolk-sac larvae of Atlantic halibut (Hippo glossus hippoglossus ). Journal of Fish
Biology, 44: 303-3 10.
Rust, M.B., Hardy, R.W. and Stickney, R.R.,1993. Assimilation of three biochemical,
forms of radio-labeled methionine in four species of fish larvae (abstract). In: M.
Carrillo L. Dahle, J. Morales, P. Sorgeloos, N. Svennevig and J. Wyban
(Compilers). From Discovery to Commercializization. European Aquaculture
Society Publication, #19: 261.
Segner, H., Roland, R., Johan, V. and Ulrich, W., 1993. Larval nutritional physiology:
Studies with Clarias gariepinus, Coregonus lavaretus and Scopthalmus maximus.
Journal of The World Aquaculture Society, 24: 121-134.
Sommerville, G. R., 1962, Southeast Research Institute, U.S.A. Patent 3,015,128.
Southgate, P. C. and Lou, D. C., 1995. Improving the n -3 HUFA composition of
Artemia using microcapsules containing marine oils. Aquaculture, 134: 9 1-99.
Steel, G. D. R. and Torrie, H. J., 1980. Principles and procedures of statistics: a
biometrical approach. McGraw-Hill mc, pp. 23 1-233.
Stepien, W. P., 1976. Feeding of laboratory-reared larvae of the seabream, Archosargus
rhomboidalis (Sparidae). Marine Biology, 38: 1-16.
Tandler, A. and Kolkovski S., 1991. Rates of ingestion and digestibility as limiting fctors
in the successful use of microdiets in Sparus aurata larval rearing. Larvi '91, EAS
Special Publication No: 15 Ghent, Belgium, pp. 169-171.
Van Der Meeren, T., 1991. Algae as first food for cod larvae, Gadus morhua L.: filter
feeding or ingestion by accident ? Journal of Fish Biology, 39: 225-237.
Villamar, D. F. and Langdon, C. J., 1993. Delivery of dietary components to larval
shrimp (Panaeus vannamei) by means of complex microcapsules. Marine Biology,
115: 635-642.
Walford, J., Lim, T. M. and Lam, T. J., 1991. Replacing live foods with
microencapsulated diets in the rearing of sea bass (Lates calcarfer) larvae: do the
larvae ingest and digest protein-membrane microcapsules? Aquaculture, 92: 225235.
Walford, J. and Lam, T. J., 1993. Development of digestive tract and proteolytic enzyme
activity in seabass (Lates calcarifer) larvae and juveniles. Aquaculture, 109: 187205.
Watanabe, Y., 1984. An ultrastructural study of intracellular digestion of horseradish
peroxidase by the rectal epithelium cells in larvae of a freshwater cottid fish Cottus
nozawae. Bulletin of The Japanese Society of Scientific Fisheries, 50: 409-4 16.
Watanabe, T., 1993. Importance of docosahexaenoic acid in marine larval fish. Journal of
the World Aquaculture Society, 24: 152-16 1.
Webster, C. D. and Lovell, R. T., 1990. Responses of striped bass larvae fed brine
shrimp from different sources containing different fatty acid compositions.
Aquaculture, 90: 49-61.
Westerfield, M., 1993. The zebrafish book. University of Oregon Press.
Weinhart, G. and Rösch, R., 1991. Food intake of larvae of Coregonus lavaretus L. Do
they really ingest less dry diet than Artemia nauplii? Page 144 in P. Lavens, P.
Sorgeloos, E. Jaspers and F. 011evier, editors. Larvi '91. Fish and Crustacean
Larviculture Symposium. EAS Special Publication, No. 15: 144.
Zambonino Infante, J.L. and Cahu, C., 1994. Development and response to a diet change
of some digestive enzymes in seabass (Dicentrarchus labrax) larvae. Fish
Physiology and Biochemistry, 12: 399-408.
81
APPENDICES
APPENDIX A
Table 1. Multifactor ANOVA table for a comparison of the effect of diet and container type
on growth of zebrafish larvae.
Source of variation
Sum of squares
d.f.
Mean square
F-ratio
Sig. level
Diet
10.998
1
10.998
197.539
0.0001
Container type
0.797
2
0.399
7.158
0.009
Diet*container type
0.276
2
0.138
2.477
0.1257
Residual
0.668
12
0.056
Total
17
Table 2. Tukey HSD comparison table for effect of diet and container type on growth of
zebrafish larvae after 8 days of feeding.
Treatment
Number of samples
MLs
Homogenous groups
Rectangular MF
3
5.32
X
Cylinder MF
3
5.34
X
ConeMF
3
5.56
X
Rectangular MF+ART
3
6.56
X
Cylinder MF+ART
3
6.96
XX
Cone MF+ART
3
7.37
X
*3
APPENDIX A (Continued)
Table 3. One way ANOVA table for differences among percent survival rates of zebrafish
larvae fed on either MF or MF+ART.
Source of variation
Sum of squares
d.f.
Mean square
F-ratio
Sig. level
Diet
0.01388
1
0.01388
5.15
0.0374
Residual
0.04311
16
0.00269
Total
0.057
17
APPENDIX B
Table 1. Simple regression table for the absorbances of different concentrations of Poly-R
in distilled water (residual df = 4).
Source of variation
Estimate
Standart error
T-statistic
P-value
Intercept
0.000345
0.000372
0.9266
0.4065
Slope
11075.3
7.989
1386.16
0.0000
Table 2. Simple regression table for the absorbances of different concentrations of Poly-R
in 1 M NaOH (residual df = 5).
Source of variation
Estimate
Standart error
T-statistic
P-value
Intercept
0.0013
0.00552
0.235275
0.8233
Slope
14010.6
64.4058
217.537
0.0000
APPENDIX C
Table 1. One way ANOVA table for gut fullness values of first feeding zebrafish larvae
3.8 mm) fed on MF, CLPWC or GABas well as starved larvae.
Source of variation
Sum of squares
d.f.
Mean square
F-ratio
Sig. level
Between groups
0.00181
3
0.0006
79.74
0.0000
Within groups
0.00006
8
0.0000
Total
0.00187
11
Table 2. Tukey HSD comparison table for effect of diet type (MF, CLPWC, GAB or
starved larvae) on gut fullness values of first feeding
3.8 mm) zebrafish larvae.
Treatment
Number of samples
Mean foregut area
Homogenous groups
Starved
3
0.083
X
CLPWC
3
0.099
X
GAB
3
0.101
X
MF
3
0.118
X
APPENDIX C (Continued)
Table 3. One way ANOVA table for feeding incidences of first feeding zebrafish larvae
3.8 mm) fed on MF, CLPWC or GAB.
(MLS
Source of variation
Sum of squares
d.f.
Mean square
F-ratio
Sig. level
Betweengroups
5616.22
2
2808.11
144.42
0.0000
Within groups
116.667
6
19.444
Total
5732.89
8
Table 4. Tukey HSD comparison table for effect of diet type (MF, CLPWC or GAB) on
feeding incidences of first feeding (MLS 3.8 mm) zebrafish larvae.
Treatment
Number of samples Mean feeding incidence
CLPWC
3
33
GAB
3
67.7
MF
3
94
Homogenous groups
X
X
X
APPENDIX D
Table 1. One way ANOVA table for gut fullness values of 12 day-old zebrafish larvae
4.2 mm) fed on MF, CLPWC or GAB as well as starved larvae.
Source of variation
Sum of squares
d.f.
Mean square
F-ratio
Sig. level
Betweengroups
0.00198
3
0.000662
38.61
0.0000
Within groups
0.00013
8
0.000017
Total
0.00021
11
Table 2. Tukey HSD comparison table for effects of diet type (MF, CLPWC, GAB or
4.2 mm) zebrafish larvae.
starved larvae) on gut fullness values of 12 day-old
Treatment
Number of samples
Mean stomach area
Homogenous groups
Starved
3
0.095
X
CLPWC
3
0.113
GAB
3
0.126
X
MF
3
0.128
X
X
nyu
APPENDIX D (Continued)
Table 3. One way ANOVA table for feeding incidences of 12 day-old zebrafish larvae
4.2 mm) fed on MF, CLPWC or GAB.
Source of variation
Sum of squares
d.f.
Mean square
F-ratio
Sig. level
Betweengroups
2184
2
1092
93.6
0.0000
Within groups
70
6
11.67
Total
2254
8
Table 4. Tukey HSD comparison table for effect of diet type (MF, CLPWC or GAB) on
feeding indidences of 12 day-old (MLS 4.2 mm) zebrafish larvae.
Treatment
Number of samples
Mean feeding incidence
Homogenous groups
CLPWC
3
62
X
GAB
3
94
X
MF
3
96
X
(MLS
APPENDIX E
Table 1. One way ANOVA table for gut fullness values of 19 day-old zebrafish larvae
5.2 mm) fed on MF, CLPWC or GAB as well as starved larvae.
Source of variation
Sum of squares
d.f.
Mean square
F-ratio
Sig. level
Between groups
0.02188
3
0.007296
515.09
0.0000
Within groups
0.0001 13
8
0.000014
Total
0.02200
11
Table 2. Tukey HSD comparison table for effects of diet type (MF, CLPWC, GAB or
starved larvae) on gut fulines values of 19 day-old (MLS 5.2 mm) zebrafish larvae.
Treatment
Number of samples
Mean stomach area
Homogenous groups
Starved
3
0.195
X
CLPWC
3
0.259
MF
3
0.284
GAB
3
0.310
X
X
X
APPENDIX E (Continued)
Table 3. One way ANOVA table for feeding incidences of 19 day-old zebrafish larvae
5.2 mm) fed on MF, CLPWC or GAB.
Source of variation
Sum of squares
d.f.
Mean square
F-ratio
Sig. level
Betweengroups
2.666
2
1.333
0.71
0.5305
Within groups
11.333
6
1.888
Total
14
8
Table 4. Tukey HSD comparison table for effect of diet type (MF, CLPWC or GAB) on
feeding incidences of 19 day-old (MLS 5.2 mm) zebrafish larvae.
Treatment
Number of samples
Mean feeding incidence
Homogenous groups
CLPWC
3
94.7
X
GAB
3
95.3
X
MF
3
96
X
(MLS
91
APPENDIX F
Table 1. One way ANOVA table for differences among cones in the growth of zebrafish
larvae (17 day-old) pre-conditioned on a diet of MF for 10 days followed by Arteinia
nauplii for 2 days.
Source of variation
Sum of squares
d.f.
Mean square
F-ratio
Sig. level
Between groups
0.008
7
0.00114
0.32
0.9339
Within groups
0.057133
16
0.0035708
Total
0.0651333
23
Table 2. One way ANOVA table for differences among cones in survival of zebrafish
larvae (17 day-old) pre-conditioned on a diet of MF for 10 days followed by Artemia
nauplii for 2 days.
Source of variation
Sum of squares
d.f.
Mean square
F-ratio
Sig. level
Betweengroups
18.625
7
2.66071
0.61
0.7368
Within groups
69.3333
16
4.33333
Total
87.9583
23
APPENDIX F (Continued)
Table 3. One way ANOVA table for differences among growth of zebrafish larvae (25 dayold) fed on Artemia nauplii substituted with different levels of MF encapsulated within
CLPWC after 8 days of feeding.
Source of variation
Sum of squares
d.f.
Mean square
F-ratio
Sig. level
Betweengroups
7.52054
6
1.25342
169.12
0.0000
Within groups
0.103762
14
0.0074115
Total
7.62431
20
Table 4. Tukey HSD comparison table for effect of substitution of Artemia nauplii with
different levels of CLPWC on growth of zebrafish larvae after 8 days of feeding.
Treatment
Number of samples
MLs
Homogenous groups
100% MF
3
5.293
X
100% CLPWC
3
5.272
X
80% CLPWC+20% ART
3
6.158
60% CLPWC+40% ART
3
6.473
40% CLPWC+60% ART
3
6.582
X
20% CLPWC+80% ART
3
6.75 9
X
100% ART
3
6.771
X
X
X
93
APPENDIX F (Continued)
Table 5. One way ANOVA table for differences among survival of zebrafish larvae (25
day-old) fed on Artemia nauplii substituted with different levels of MF encapsulated within
CLPWC after 8 days of feeding.
Source of variation
Sum of squares
d.f.
Mean square
F-ratio
Sig. level
Between groups
712.778
6
118.796
90.69
0.0000
Within groups
18.3388
14
1.30991
Total
731.117
20
Table 6. Tukey HSD comparison table for effect of substitution of Artemia nauplii with
different levels of CLPWC on survival of zebrafish larvae after 8 days of feeding.
Treatment
Number of samples
Mean survival
Homogenous groups
100%CLPWC
3
43
X
80% CLPWC+20% ART
3
64.3
X
60% CLPWC+40% ART
3
68
XX
40% CLPWC+60% ART
3
69.6
XX
100% MF
3
70.3
XX
20% CLPWC+80% ART
3
72.7
XX
100% ART
3
74.3
X
APPENDIX G
Table 1. One way ANOVA table for differences among cones in the growth of zebrafish
larvae (17 day-old) pre-conditioned on a diet of MF for 10 days followed by Artemia
nauplii for 2 days.
Source of variation
Sum of squares
d.f.
Mean square
F-ratio
Sig. level
Between groups
0.0092
7
0.001314
0.89
0.5392
Within groups
0.0237
16
0.001483
Total
0.0329
23
Table 2. One way ANOVA table for differences among cones in the survival of zebrafish
larvae (17 day-old) pre-conditioned on a diet of MF for 10 days followed by Arternia
nauplii for 2 days.
Source of variation
Sum of squares
d.f.
Mean square
F-ratio
Sig. level
Betweengroups
18.625
7
2.66071
0.61
0.7368
Within groups
69.3333
16
4.33333
Total
87.9583
23
95
APPENDIX G (Continued)
Table 3. One way ANOVA table for differences among growth of zebrafish larvae (25 dayold) fed on Arteinia nauplii substituted with different levels of MF encapsulated within
GAB after 8 days of feeding.
Source of variation
Sum of squares
d.f.
Mean square
F-ratio
Sig. level
Between groups
7.91 891
6
1.31982
115.35
0.0000
Within groups
0.160186
14
0.01144
Total
8.0791
20
Table 4. Tukey HSD comparison table for effect of substitution of Arternia nauplii with
different levels of GAB on growth of zebrafish larvae after 8 days of feeding.
Number of samples
MLs
100% MF
3
6.196
100% GAB
3
5.873
80% GAB+20% ART
3
6.268
60% GAB+40% ART
3
6.633
40% GAB+60% ART
3
7.185
X
20% GAB+80% ART
3
7.377
XX
100% ART
3
7.608
X
Treatment
Homogenous groups
X
X
X
X
APPENDIX G (Continued)
Table 5. One way ANOVA table for differences among survival of zebrafish larvae (25
day-old) fed on Artemia nauplii substituted with different levels of MF encapsulated within
GAB after 8 days of feeding.
Source of variation
Sum of squares
d.f.
Mean square
F-ratio
Sig. level
Between groups
0.00909
6
0.001515
5.05
0.0059
Within groups
0.0042
14
0.0003
Total
0.013295
20
Table 6. Tukey HSD comparison table for effect of substitution of Arternia nauplii with
different levels of GAB on survival of zebrafish larvae after 8 days of feeding.
Treatment
Number of samples
Mean survival
Homogenous groups
100% GAB
3
92.3
X
60% GAB+40% ART
3
94.3
XX
100% ART
3
94.7
XX
100% MF
3
94.7
XX
20% GAB+80% ART
3
96.7
XX
80% GAB+20% ART
3
98
X
40% GAB+60% ART
3
98.7
X